Macquarie Technicals PDF

Quantitative analysis
AUSTRALIA
Lets get technical
30 April 2009
Technical indicators add value in Australia

Technical indicators appear to have predictive power in Australian markets.
These signals are typically a more sophisticated form of short term mean
reversion signals. We think these results are encouraging as the standard one
month mean reversion signal has typically not been predictive in Australia.
Volatile markets present opportunity to use technicals

The large macro-driven swings in equity markets have resulted in deterioration of
performance for a number of elementary quantitative factors. The factor volatility
makes it necessary to utilise smarter techniques when searching for Alpha. A
large number of technical indicators have had strong performance in the volatile
period since August 2007.
Williams %R the most powerful signal

The Williams %R signal stands out as the strongest performing technical
indicator, with an IC 5.6%. The Williams %R is an enhanced mean reversion
signal and measures the relative location of the closing price in relation to a
recent high-low range. Other key indicators that worked well were: force index,
commodity channel index and relative strength indicator.
Size matters technicals work best in large cap stocks

Inside
What goes up must come down
Market instability hurts fundamental

factors
Using Technical Indicators as Quant

factors
In doing this analysis, we also found that there was a significant difference in
behaviour between large cap and small cap stocks.
Large cap stocks (ASX100) tend to be mean reverting which favours the
mean reverting technical indicators over trend following strategies.

Small cap stocks (ex-ASX100) tend to be trend following, this penalises the
Performance of Technical Indicators in

Australia
5
Combining with other factors and use in
a factor model
10
References
14
Appendices Factor definitions
15
mean trend Technical indicators and favours momentum based strategies.
Improving the investment process with technicals

Incorporating a technical trading signal in an existing investment process can
add significant value as these signals tend to be lowly correlated to the standard
value, momentum, quality, growth and analyst sentiment strategies. However,
careful consideration must be given to the higher turnover of these signals which
makes them hard to include in a process.
We look at different ways of incorporating the technical indicators in an
investment process. To do this we use the Williams %R:
as an extra factor in a standard four factor model; and
Analysts
Burke Lau
612 8232 0481
John Conomos
612 8232 5157
George Platt
612 8232 6539
as a knock out signal in a standard quant model.

burke.lau@macquarie.com
john.conomos@macquarie.com
In general, we find that the knock out method was a good way to incorporate
the predictive strength and timing ability of the technical indicators while
moderating the effects of higher turnover.
george.platt@macquarie.com
Please refer to the important disclosures and analyst certification on inside back cover of
this document, or on our website www.macquarie.com.au/research/disclosures.
Macquarie Research Equities - Report
Lets get technical

Short term technical
indicators are
typically enhanced
mean reverting
signals.
Defining technical indicators

With the recent volatility in the market as well as weakness in the performance of some of the
standard quant factors like value, we turn our attention to faster burn short term strategies
that may work better in volatile markets.
Technical indicators are generally short term mean reverting strategies. In this report, we
examine a number of technical indicators in the Australian market and test their efficacy in
forecasting short term stock performance.
Williams %R the standout technical indicator

Williams %R was the standout technical indicator from our analysis. It has had consistent
performance through time and across the large cap and small cap stocks.
Fig 1 Williams %R information coefficient
Fig 2 Technical performance in large vs. small caps
40% Information
Coefficient
30%
Macquarie Alpha
Model
12M Momentum
20%
3M Earnings
Revisions
10%
Earnings Yield
0%
Price relative to
52 Week High
-10%
Force Index
-20%
-30%
De
c
Ju 94
Fe l 95
b
Se 96
p
Ap 96
Nor 97
v
Ju 97
n
Ja 98
n
Au 99
g
M 99
ar
O 00
c
M t 00
ay
De 01
c
Ju 01
Fe l 02
b
Se 03
p
Ap 03
Nor 04
v
Ju 04
n
Ja 05
n
Au 06
g
M 06
a
O r 07
c
M t 07
ay
De 08
c
08
-40%
Monthly IC
12M Average IC
All
Last 5
Last 3
Last 1
History Years
Years
Year
Average IC
5.6%
5.4%
4.8%
5.9%
Std Deviation IC
9.6%
7.9%
8.7%
12.3%
Success Rate
73.7%
73.3%
63.9%
50.0%
Avg IC / Std Dev IC
0.59
0.68
0.55
0.48
t-stat
7.68
5.29
3.29
1.65
Technical Indicators
Williams %R
Bollinger Band
Commodity
Channel Index
Relative Strength
Indicator
Money Flow Index
Stochastic
Oscillator
On Balance
Volume
1M Mean
Reversion
Small Cap - ex ASX100

Large Cap - ASX100
MACD
-4%
-2%
0%
2%
4%
6%
8%
10%
12%
Information Coefficient
Source: Macquarie Research, April 2009
Other key technical indicators that also performed well were:
perform strongly in
the large cap space.
Force Index
Bollinger Band
Commodity Channel Index
Relative Strength Indicator
Large caps mean revert; small caps trend follow

Mean reverting technical indicators work significantly better in large cap stocks. The standard
one month mean reversion signal is negatively predictive in the Australian market
(IC = -0.72%). But when split by size it outperforms in the large caps (IC=1.5%) while
underperforming strongly in the small cap stocks (IC=-2.24%).
Adding value to a quant process

Typically, mean reverting technical indicators are negatively correlated with the standard
momentum, value and analyst revisions signals. This low correlation indicates that technical
indicators can add significant value to an existing process.
A drawback of the typical technical indicators is that they have very high turnover levels. The
Williams %R has an annual Q1-Q5 2-way turnover of 3835%. This is significantly higher than
typical quant factors. The increased turnover from adding a technical signal to a standard
quant process destroys most of the added alpha from the technical indicator.
An alternative is to use the technicals as a knock out overlay to a standard quant model. This
technique can take advantage of the timing ability of the technical indicators without
significantly increasing the turnover of the portfolio.
30 April 2009
Market instability hurts fundamental factors

Higher time series
and cross sectional
volatility has
increased returns to
mean reveting
strategies.
Equity markets have proved extremely challenging for investors since the foundations of the
market began to shake in the second half of 2007. As soaring levels of uncertainty persisted,
fundamental factors entrenched in security valuation have had highly volatile performance.
This volatility has made life exceptionally difficult for investors as strong performance for a
month or two is often followed by severe underperformance.
Adding to the challenges for investors is the terrible performance of typical medium to longer
term quant strategies such as value. Value factors are a vital input into most quantitative
models and these factors have had arguably their worst performance in recent history
(Figures 3 and 4).
Fig 3 ASX200 cross sectional volatility
Fig 4
Performance of value factors
%
25
50% Information
Coefficient
40%
20
30%
20%
15
10%
10
0%
-10%
-20%
Source:Macquarie Research, April 2009
-30%
19
80
19
82
19
84
19
86
19
88
19
90
19
92
19
94
19
96
19
98
20
00
20
02
20
04
20
06
20
08
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
Turning our attention to technicals

While in the long run we believe volatility will reside and the fundamental factors will again
return to the fore; in the short-term, the degree of macro instability indicates that the volatility
will persist. With this in mind we turn our attention to short term focused technical indicators.
Our European team first looked at technical factors in its March Quantamentals report1.
They tested (in European markets) whether technical factors can systematically add value on
a standalone basis and if so, can they be incorporated into the investment process. We focus
specifically on Australia to see how the results compare.
Technical analysis is a contentious topic, with many investors and members of the academic
community expressing strong opinions on the validity of the signals. This often leads to
heated debate with critics rejecting the use of these metrics. Technical analysis is often
viewed as being too subjective as technical analysts will interpret the same chart in several
different ways depending upon the metric employed and also the period being examined.
Hence, investors typically do not include these metrics within their models.
30 April 2009
Davies, C. Quantamentals: Do technicals add value?, Macquarie Research Equities, (2009)
Using technical indicators as Quant factors

The number of technical indicators in existence is significant, so we have mainly focused on
the most commonly available indicators. These include the more frequently discussed
indicators on websites and those most readily available on platforms such as Factset and
Bloomberg. Figure 5 below shows the indicators that we tested. A brief description and
definitions of these indicators are included in the Appendix.
Fig 5
Technical indicators tested
Price Based Technical Indicators
Price & volume Based Technical Indicators
Volume Based Technical Indicators
Relative Strength Index

Williams %R
MACD
Stochastic Oscillator
Bollinger Band %b
Moving Averages
Price Relative to 52 Week High & Low
Average Directional Movement Index (ADX)
Relative Volatility
Money Flow Index

Chaikin Money Flow Index
On Balance Volume
Force Index
Accumulation / Distribution Index
Trading Intensity
We test the standard

signals
When testing the technical indicators as quant factors, a key characteristic to keep in mind is
that most of the indicators were designed to be used on a univariate basis, relative to its own
history. Typically, these technical indicators attempt to detect when a stock (relative to its
own history):
is over-bought
is over-sold
has broken out from a trading range
changes its price trends
Because of this, the scale of some of the technical indicators may not match up when
compared cross-sectionally.
We time series
normalise technical
indicators to detect
changes in the
signal relative to its
own stock history.
This is different from typical quant signals which tend to be used cross-sectionally to compare
and rank one stock against another, eg on balance volume is a running total of (volume *
direction of price move), which typically grows over time is very different for large cap vs.
small cap stocks.
To take this into account we normalise a signal against its own history (using the last 12
months of data). This makes the signals:
comparable in magnitude,
picks up any changes relative to its own history and
filters out stock and market drift effects (momentum) in the signal.
Additionally, the majority of the technical indicators use a trailing window of data to calculate
the underlying signal. We notice from our backtests that a window of ten trading days typically
yields the best results. Given this result we will be using the 10-day-based metrics from this
point forward rather than spend hours debating whether we should a 10-day, 14-day or 11day window.
30 April 2009
Performance of technical indicators in

Australia
Fig 6 Technical factor performance in Australia Dec 1994 Mar 2009
Average IC
Q1-Q5 Long Short
IC Avg/Sd IC Success
t-stat
Rate
Price Momentum
3 Month Momentum
5.5%
1.47
6 Month Momentum
5.7%
1.45
12 Month Momentum
7.1%
1.70
Price Reversal
1 Month Mean Reversion
-0.7%
-0.21
Relative Strength Indicator - 14 Day (Inverted)
Standard
1.3%
0.42
12M Normalised
3.6%
1.33
Williams %R (Inverted)
Standard
4.2%
1.56
12M Normalised
5.6%
2.22
52 Week High / Low
Price relative to 52 Week High
6.6%
1.57
Price relative to 52 Week Low
2.7%
0.69
Bollinger Band %b (Inverted)
%b
2.7%
0.89
%b - 12 Month Normalised
4.4%
1.67
Force Index (inverted) - 10 Period
Standard
3.8%
1.66
12 Month Normalised
4.4%
1.94
Commodity Channel Index - 10 Period (Inverted)
Standard
2.7%
1.00
12 Month Normalised
4.1%
1.69
Stochastic Oscillator (Inverted)
Standard
-2.0%
-0.62
12M Normalised
1.0%
0.34
MACD (Inverted)
12 Month Normalised
-0.8%
-0.27
Money Flow Index (Inverted)
Standard
0.7%
0.28
12 Month Normalised
2.1%
0.92
Moving Average (20 Period / 60 Period)
Standard
2.4%
0.56
12 Month Normalised
1.2%
0.45
Standard
0.2%
0.07
12 Month Normalised
-1.0%
-0.41
Average Directional Movemebnt Index (ADX)
Standard
-0.2%
-0.09
12 Month Normalised
-0.8%
-0.43
Average True Range (Inverted) - 10 Period
Standard
1.7%
0.55
12 Month Normalised
-0.7%
-0.26
Chankin Money Flow Index - 21 Period
Standard
0.3%
0.11
12 Month Normalised
-1.1%
-0.42
On Balance Volume
12 Month Normalised
0.7%
0.31
Trading Intensity
1M / 12M Average Daily Volume 4.7%
1.93
Relative Volatility
1M / 12M Daily Volatility
2.2%
0.95
Return
(p.a)
Volatility Informatio
(p.a.)
n Ratio
Success
Rate
Largest 1
Month
Gain
Largest 1
Month
Loss
Turnover
(2 Way
p.a.)
67.8%
68.4%
71.3%
29.1%
27.6%
29.3%
17.7%
19.8%
19.9%
1.65
1.40
1.48
73.1%
73.1%
73.1%
19.2%
16.4%
16.2%
-23.7%
-31.1%
-29.5%
2076%
1476%
1062%
46.8%
-10.5%
14.8%
-0.71
43.3%
15.5%
-16.1%
3655%
51.5%
63.7%
-4.5%
5.9%
12.1%
10.2%
-0.37
0.58
47.4%
58.5%
12.7%
12.0%
-11.4%
-9.3%
3808%
3888%
69.6%
73.7%
7.5%
13.2%
10.0%
10.2%
0.75
1.29
63.7%
66.7%
11.5%
12.3%
-9.7%
-7.5%
3776%
3835%
73.1%
62.0%
24.6%
15.9%
18.3%
17.0%
1.34
0.94
69.0%
66.7%
20.7%
15.4%
-30.0%
-28.2%
1767%
1363%
60.8%
67.3%
0.6%
7.7%
10.5%
10.3%
0.06
0.76
56.7%
64.9%
11.7%
10.9%
-8.3%
-7.3%
3847%
3886%
67.3%
69.0%
6.8%
9.4%
9.2%
8.8%
0.73
1.07
60.8%
64.9%
11.0%
9.9%
-10.9%
-11.7%
3332%
3717%
59.1%
65.5%
-0.1%
8.0%
10.1%
9.1%
-0.01
0.88
50.9%
63.2%
10.4%
12.9%
-9.9%
-6.8%
3830%
3865%
44.4%
50.9%
-10.5%
0.9%
12.9%
10.7%
-0.81
0.08
39.8%
52.6%
17.1%
12.5%
-12.8%
-10.5%
3588%
3654%
48.5%
-7.4%
11.9%
-0.63
44.4%
12.8%
-11.2%
3395%
52.6%
62.6%
-4.4%
0.8%
9.1%
8.5%
-0.48
0.09
38.0%
50.9%
12.7%
10.9%
-8.1%
-7.1%
3758%
3837%
52.0%
57.9%
-1.3%
1.9%
19.7%
10.7%
-0.07
0.17
45.0%
55.6%
30.1%
12.4%
-16.7%
-12.5%
2036%
2872%
50.3%
50.3%
1.9%
-1.1%
11.5%
9.8%
0.16
-0.12
51.5%
49.1%
13.1%
9.5%
-11.7%
-10.0%
3076%
3716%
46.8%
47.4%
-1.0%
-2.5%
7.3%
7.1%
-0.13
-0.36
49.1%
43.9%
7.8%
6.5%
-6.3%
-6.4%
3877%
3711%
52.0%
42.1%
3.9%
-5.6%
12.3%
11.8%
0.31
-0.48
59.1%
41.5%
19.8%
22.6%
-13.5%
-11.3%
1234%
3302%
52.6%
48.0%
2.4%
-1.1%
11.1%
9.8%
0.22
-0.11
54.4%
54.4%
8.5%
7.5%
-12.0%
-11.7%
2829%
3661%
51.5%
-2.3%
9.5%
-0.24
41.5%
11.4%
-7.4%
3761%
70.2%
18.1%
11.7%
1.55
68.4%
9.2%
-16.8%
2723%
60.8%
6.8%
9.7%
0.70
62.6%
8.6%
-12.1%
3416%
30 April 2009
Momentum
strategies are
strongly predictive
in Australia
A key characteristic of the Australian market is the stellar performance of the 12 month price
momentum factor. This is where stocks with high (low) 12-month price momentum continue
to outperform (underperform). The intuition behind the momentum phenomenon in Australia
resides in the recurrent cash injections from compulsory superannuation. Managers continue
to invest the new funds into their preferred holdings, which in turn drives the performance of
momentum.
On the flipside of 12 month momentum is the mean reversion effect. To measure this, we
historically use 1 month mean reversion, which simply means stocks that have outperformed
over the past month are likely to underperform in the current month. The underlying logic
behind many technical indicators is some form of mean reversion. A number of the indicators
are effectively enhanced reversion signals.
Figure 6 above shows the overall performance of a number of momentum and technical
signals in the Australian market from December 1994 to April 2009. A key result is that the
standard short term mean reversion signal of 1M mean reversion does not seem to work in
Australia, with a negative IC of -0.8% and a long/short quintile return of -10.4% pa.
Surprisingly, a number of the short term technical indicators which are mean reverting
strategies seemed to perform much better than the standard 1 month mean reversion signal
and actually were strongly predictive.
Simple 1 month
mean reversion
doesnt work in
Australia
The key mean reverting technical indicators which added value in the Australian market are
shown in Figure 7.
Fig 7
Information coefficient of mean reverting technical indicators

Williams %R
Force Index
Bollinger Band
Commodity
Channel Index
Relative Strength
Indicator
Money Flow Index
Stochastic
Oscillator
On Balance
Volume
1M Mean
Reversion
-2%
-1%
0%
1%
2%
3%
4%
5%
6%
30 April 2009
Williams %R stands
out from the crowd
The Williams %R signal stands out from this list of technical indicators. It has a strong
consistent IC as well as good quintile long/short performance. Also encouraging with this
signal is its consistent performance in most developed markets2. The charts below highlight
some performance statistics for the signal.
Fig 8 Williams %R information coefficient

40%
30%
Fig 9 Q1-Q5 long / short performance

50%
Information
Coefficient
40%
20%
Long Short
Returns
30%
10%
0%
20%
-10%
10%
-20%
0%
-30%
De
c
Ju 94
Fe l 9
5
Seb 96
p
Ap 96
Nor 97
Juv 97
n
Ja 98
Aun 9
g 9
M 99
ar
O 00
c
M t0
a 0
Dey 01
c
Ju 01
Fe l 0
b 2
Se 03
p
Ap 03
Nor 04
Juv 04
n
Ja 05
Aun 0
6
Mg0
ar 6
O 07
c
M t0
a 7
Dey 08
c
08
Monthly IC
12M Average IC
All History
Average IC
Std Deviation IC
Success Rate
Avg IC / Std Dev IC
t-stat
5.6%
9.6%
73.7%
0.59
7.68
Last 5
Years
5.4%
7.9%
73.3%
0.68
5.29
Last 3
Last 1 Year
Years
4.8%
5.9%
8.7%
12.3%
63.9%
50.0%
0.55
0.48
3.29
1.65
De
c
J 94
Feul 9
5
Seb 96
p
Ap 96
Nor 9
7
Juv 9
7
Jan 9
8
Aun 9
9
Mg9
ar 9
O 00
Mct 0
a 0
Dey 01
c
J 01
Feul 0
2
Seb 0
p 3
Ap 03
Nor 0
4
Juv 0
4
n
Ja 0
5
Aun 0
6
Mg0
a 6
O r 07
Mct 0
a 7
Dey 0
c 8
08
-10%
-40%
LS Return
12M Rolling LS returns
Last 5
Last 3
Last Year
Years
Years
Quartile 1 Excess Return *
4.4%
3.8%
3.5%
3.7%
Quartile 5 Excess Return *
-8.3%
-10.4%
-11.9%
-23.3%
Q1 - Q5 Return (p.a.) #
13.2%
15.1%
16.4%
32.1%
Q1 - Q5 Volatility (p.a.) #
10.2%
10.1%
12.2%
19.6%
Q1 - Q5 Information Ratio #
1.29
1.50
1.34
1.64
Q1 - Q5 Turnover 1 way p.a.#
3841%
3868%
3844%
3867%
* Monthly rebalanced annualised excess returns (long - market and market - short)
# Monthly rebalanced Long Q1 - Short Q5
All History
Of particular interested is the recent pick-up in the predictive strength of the Williams %R
signal which is in line with a pick up in performance of a number of technical indicators we
have noticed over the last year.
30 April 2009
Davies, C. Quantamentals: Do Technicals add value?, Macquarie Research Equities, (2009)
Size performance
In our report, dissecting the Alpha model in Australia3, we further examined factor
performance across different universes. One of the key findings was the shifting performance
of factors across the different universes definitions. To test this we divided the universe into
two:
A large cap universe containing the top 100 stocks in the market and
A small cap universe consisting of the top 300 stocks ex the top 100 stocks.
Fig 10 Technical indicator information coefficient in

large vs. small cap universes
Fig 11 Q1-Q3 L/S information ratio before costs in

large vs small cap universe
Macquarie Alpha
Model
Macquarie Alpha
Model
12M Momentum
12M Momentum
3M Earnings
Revisions
3M Earnings
Revisions
Earnings Yield
Earnings Yield
Price relative to
52 Week High
Price relative to
52 Week High
Force Index
Force Index
Williams %R
Bollinger Band
TechnicalIndicators
Williams %R
Commodity
Channel Index
Relative Strength
Indicator
Money Flow Index
Stochastic
Oscillator
On Balance
Volume
1M Mean
Reversion

Large Cap - ASX100
MACD
Bollinger Band
Commodity
Channel Index
Relative Strength
Indicator
Money Flow Index
Stochastic
Oscillator
On Balance
Volume
1M Mean
Reversion

Large Cap - ASX100
MACD
-4%
-2%
0%
2%
4%
6%
8%
10%
12%
-1.50
-1.00
-0.50
0.00
0.50
1.00
1.50
2.00
Information Ratio
Technical indicators
perform much
stronger in the large
cap markets
The key result above is the significant difference in performance between the large cap and
small cap universe for the technical indicators. The majority of the technical indicators are
mean reverting signals, and perform significantly better in the large cap universe. While the
more traditional momentum signals like 12M momentum and earnings revisions favour the
smaller cap universe.
30 April 2009
R. De Souza, Macquarie Alpha Model The alpha model goes under the knife
8
Turnover
A common criticism of technical indicators is the significant increase in turnover. Figure 12
highlights this by showing the average monthly two-way turnover on an annualised basis for a
variety of factors. This is based upon a long short Q1-Q3 strategy over the ASX100 universe.
For the more common Quant factors the turnover ranges from 600% to 1,200%. The turnover
for the technical indicators is much higher with all the mean reverting technical factors
measuring over 3,000%4.
Fig 12 Technical indicator Q1-Q3 L/S annual turnover

in ASX 100 universe
Fig 13 Q1-Q3 L/S after cost information ratio in ASX

100 universe with 15bp and 30bp transaction costs
Earnings Yield
Earnings Yield
12M Momentum
12M Momentum
Macquarie Alpha Model
Macquarie Alpha Model
3M Earnings
Revisions
Price relative to
52 Week High
3M Earnings
Revisions
Price relative to
52 Week High
Trading Intensity
Trading Intensity
1M Mean
Reversion
1M Mean
Reversion
After Cost IR (30bps)

After Cost IR (15bps)
MACD
MACD
On Balance
Volume
Stochastic
Oscillator
On Balance
Volume
Stochastic
Oscillator
Money Flow Index
Money Flow Index
Force Index
Force Index
Relative Strength
Indicator
Commodity
Channel Index
Relative Strength
Indicator
Commodity
Channel Index
Williams %R
Williams %R
Bollinger Band
Bollinger Band
0%
-1.00
500% 1000% 1500% 2000% 2500% 3000% 3500% 4000%

Average Annual 2 way Turnover
have very high
turnovers which can
impact the Alpha
available from the
signals
Before Cost IR
-0.50
0.00
0.50
1.00
Average Information Ratio
1.50
To further examine the real effects of turnover, we have calculated the information ratios for
the same set of factors both pre and post transaction costs.5 The results are illustrated in
Figure 13. The high information ratio (even after costs) for a number of the other technical
factors (Williams %R, Bollinger Bands, Commodity Channel, RSI and Force Index) highlights
the potential for technical indicators to add value in the investment process.
Of the technical factors Williams %R stands out overall:
it has a strong information ratio after costs,
performs consistently through time and
works well in both the large cap and small cap space.
Highest possible turnover for a long short quintile strategy is 4800% p.a. or 400% of the face value per month
(200% long, 200% short). If the thirds are assumed to be randomly allocated then we would expect each third
to only have 1/3 in common with the previous month, so the expected average turnover for a random long short
portfolio is 3168% pa. So the mean reverting technical indicators get very close to a random portfolio in
turnover terms, with near zero auto correlation in stocks held.
5
To calculate the post transaction cost information ratio, we take the simple long-short return for a particular
month and then corresponding monthly two-way turnover multiplied by the assumed transaction cost of 30bp.
This gives a monthly long-short return series adjusted for transaction costs from which we can calculate a post
transaction cost turnover.
30 April 2009
Combining with other factors and use in a

factor model
Given the encouraging results of the univariate tests, we examine ways that investors may be
able to implement the signals into their investment process.
Correlations with other factors

A key consideration in deciding whether to use the technical indicators in an existing process
is how correlated these signals are to the existing signals in the model. The more uncorrelated the strategies the more value they will add to the process. Figure 14 below shows
the IC correlation matrix for the different factors from December 1994 to March 2009. For
comparison, we included in this analysis the Macquarie Alpha model as a representative
quant model as well as some standard quant signals:
1M mean reversion,
12 month momentum,
Earnings revisions and
Earnings yield.
Fig 14 IC Correlation matrix for quant factors and technical indicators
88%
68%
78%
64%
50%
59%
12%
31%
31%
-4%
-36%
88%
100%
59%
78%
51%
41%
56%
16%
39%
39%
-3%
-37%
74%
68%
59%
100%
28%
42%
33%
35%
13%
16%
18%
2%
-36%
5%
-14%
-41%
-20%
-25%
-25%
-28%
-19%
-27%
-21%
-49%
-44%
69%
78%
78%
28%
100%
59%
51%
49%
15%
31%
30%
-9%
-27%
-18%
-12%
-53%
-23%
-36%
-33%
-37%
-36%
-39%
-40%
-56%
-62%
72%
64%
51%
42%
59%
100%
56%
41%
11%
6%
9%
-6%
8%
10%
1%
-41%
-3%
-10%
0%
-5%
3%
-4%
-1%
-26%
-26%
56%
50%
41%
33%
51%
56%
100%
29%
4%
4%
4%
-7%
-12%
-1%
-5%
-39%
2%
-1%
0%
-6%
-2%
-9%
-2%
-25%
-23%
Trading Intensity
52%
59%
56%
35%
49%
41%
29%
100%
-7%
16%
15%
5%
-30%
-25%
-2%
-46%
-10%
-19%
-22%
-19%
-22%
-16%
-24%
-37%
-43%
Relative Volatility
18%
12%
16%
13%
15%
11%
4%
-7%
100%
16%
17%
-5%
1%
39%
-6%
42%
-1%
-8%
2%
-5%
-10%
-11%
-8%
-13%
-12%
16%
31%
39%
16%
31%
6%
4%
16%
16%
100%
98%
-3%
-12%
-31%
-3%
-15%
-21%
-28%
-39%
-38%
-53%
-39%
-51%
-41%
-44%
16%
31%
39%
18%
30%
9%
4%
15%
17%
98%
100%
-4%
-10%
-29%
-2%
-14%
-22%
-28%
-36%
-38%
-53%
-39%
-50%
-42%
-44%
-8%
-4%
-3%
2%
-9%
-6%
-7%
5%
-5%
-3%
-4%
100%
-5%
-8%
-12%
-2%
-18%
-12%
6%
-18%
9%
-8%
0%
-1%
-4%
85%
3M Momentum
72%
Price relative to
52 Week Low
Price relative to
52 Week High
Maquarie Alpha
Model
3M Earnings
Revisions
Earnings Yield
Moving Average
Force Index
Average
True Range
Commodity
Channel Index
1M Mean
Reversion
100%
6M Momentum
MACD
Earnings Yield
-30%
Stochastic
Oscillator
Average
Directional
Movement
-8%
Relative Strength
Indicator
Chaikin
Money Flow
16%
On Balance
Volume
Accumulation /
Distribution Index
16%
Bollinger Band
Relative Volatility
18%
Money
Flow Index
Trading Intensity
52%
Williams %R
3M Earnings
Revisions
56%
Commodity
Channel Index
Maquarie Alpha
Model
72%
Average
True Range
Price relative to
52 Week High
69%
Force Index
Price relative to
52 Week Low
74%
100%
Moving Average
3M Momentum
72%
12M Momentum
85%
12M Momentum
Accumulation /
Distribution Index
Chaikin
Money Flow
Average
Directional
Mean Reverting
Mean Reverting
6M Momentum
Trend Following
Trend Following
4%
-12%
-54%
-13%
-23%
-19%
-20%
-15%
-19%
-16%
-46%
-48%
-13%
-8%
-62%
-22%
-31%
-30%
-32%
-28%
-32%
-31%
-55%
-60%
-27%
-12%
-57%
-27%
-36%
-41%
-42%
-43%
-47%
-46%
-62%
-69%
-30%
-36%
-37%
-36%
-27%
8%
-12%
-30%
1%
-12%
-10%
-5%
100%
13%
4%
39%
4%
10%
11%
13%
18%
12%
18%
30%
35%
4%
-13%
-27%
5%
-18%
10%
-1%
-25%
39%
-31%
-29%
-8%
13%
100%
-8%
40%
3%
7%
20%
13%
22%
16%
40%
20%
26%
-12%
-8%
-12%
-14%
-12%
1%
-5%
-2%
-6%
-3%
-2%
-12%
4%
-8%
100%
-12%
45%
47%
24%
40%
15%
25%
-2%
6%
14%
-54%
-62%
-57%
-41%
-53%
-41%
-39%
-46%
42%
-15%
-14%
-2%
39%
40%
-12%
100%
3%
5%
13%
12%
10%
14%
24%
37%
40%
-13%
-22%
-27%
-20%
-23%
-3%
2%
-10%
-1%
-21%
-22%
-18%
4%
3%
45%
3%
100%
89%
58%
84%
49%
65%
47%
42%
46%
Williams %R
-23%
-31%
-36%
-25%
-36%
-10%
-1%
-19%
-8%
-28%
-28%
-12%
10%
7%
47%
5%
89%
100%
64%
86%
50%
70%
52%
50%
51%
Money
Flow Index
-19%
-30%
-41%
-25%
-33%
0%
0%
-22%
2%
-39%
-36%
6%
11%
20%
24%
13%
58%
64%
100%
72%
71%
75%
61%
59%
53%
Bollinger Band
-20%
-32%
-42%
-28%
-37%
-5%
-6%
-19%
-5%
-38%
-38%
-18%
13%
13%
40%
12%
84%
86%
72%
100%
67%
89%
72%
71%
68%
-15%
-28%
-43%
-19%
-36%
3%
-2%
-22%
-10%
-53%
-53%
9%
18%
22%
15%
10%
49%
50%
71%
67%
100%
71%
73%
67%
72%
-19%
-32%
-47%
-27%
-39%
-4%
-9%
-16%
-11%
-39%
-39%
-8%
12%
16%
25%
14%
65%
70%
75%
89%
71%
100%
77%
77%
72%
-16%
-31%
-46%
-21%
-40%
-1%
-2%
-24%
-8%
-51%
-50%
0%
18%
40%
-2%
24%
47%
52%
61%
72%
73%
77%
100%
80%
79%
On Balance
Volume
Relative Strength
Indicator
Stochastic
Oscillator
MACD
-46%
-55%
-62%
-49%
-56%
-26%
-25%
-37%
-13%
-41%
-42%
-1%
30%
20%
6%
37%
42%
50%
59%
71%
67%
77%
80%
100%
89%
1M Mean
Reversion
-48%
-60%
-69%
-44%
-62%
-26%
-23%
-43%
-12%
-44%
-44%
-4%
35%
26%
14%
40%
46%
51%
53%
68%
72%
72%
79%
89%
100%
The factors in figure 14 are sorted from trend following (12 month momentum) to mean
reverting (1 month mean reversion).
30 April 2009
10
Most technical factors tend to be mean reverting signals. These technical indicators are
typically an enhanced version of the standard one month mean reversion signal picking up
when a stock has been oversold or overbought. A few of the technical indicators fall into the
category of trend following signals - picking when a stock has broken out of a range, (eg
trading intensity, accumulation / distribution index or price relative to 52 week highs).
are lowly correlated
to the standard
quant factors.
A nice result is that the key predictive signals are not highly correlated to the standard quant
factors of momentum, value and analyst sentiment. The key mean reverting technical
indicators are also generally negatively correlated with the Macquarie Alpha model (Williams
%R is -10% correlated).
Even though the technical indicators are highly correlated to 1 month mean reversion they
tend to significantly outperform the standard 1M momentum measure and could be a good
substitute for this signal.
The low correlations of the technical indicators with the standard quant factors as well as
strong predictive strength indicate that it could add significant value to a quant screen or a
multi-factor composite quant model.
Combining in a multifactor model

To test how a technical indicator would perform in a multi-factor model we created a typical
equal weighted four factor model of value, revisions, quality and price momentum6 and
compared this to a new model which also includes the Williams %R factor.
in a multi factor
model struggle to
add value due to
higher turnover
Figure 15 plots the two strategies. We tested this in the large cap universe of the MSCI
Australia, where the technical indicators would add the most value and be the least impacted
by turnover constraints. Due to the smaller restricted universe we used a Q1-Q3 Long Short
portfolio to measure the performance characteristics of these portfolios.
As can be seen, replacing the 1-month reversal term with Williams %R metric does enhance
the overall strategy. From an information ratio point of view we see the information ratio jump
from 0.65 to 0.92 for an equal weighted long-short basket assuming no transaction costs.
Once transaction costs are taken into account (30bp) the information ratio increases only
from 0.30 to 0.39. Most of the performance has been eroded by the significantly higher
turnover. Figure 16 shows that the turnover has increased from 100% 2-way per month to
148% per month.
Fig 15 Standard and enhanced quant model

performance Q1-Q3 performance
Fig 16 Monthly Q1-Q5 2-way turnover
600
160%
Standard Model
Standard Model With Williams
500
140%
Standard Model (After Costs)

Standard Model With Williams (After Costs)
120%
400
100%
300
80%
200
60%
100
40%
20%
Source: FactSet, Macquarie Research, March 2009
Dec-08
Dec-07
Dec-06
Dec-05
Dec-04
Dec-03
Dec-02
Dec-01
Dec-00
Dec-99
Dec-98
Dec-97
Dec-96
Dec-95
Dec-94
Dec-93
0%
Standard Model
Source: FactSet, Macquarie Research, March 2009
We can see that using the Williams %R factor enhances returns both before costs and after
costs. However, given the high turnover nature of the Williams %R metric it is hard to include
it in the model at a significant weight.
6
For value, we use an equally weighted composite of our fundamental price to earnings and PEG ratio. For
price momentum we use 12M momentum. For Revisions we set of revisions metrics (equally weight earnings,
dividend, sales, cashflow and book). Finally, for our composite quality metric we equally weight Merton, quick
ratio, dividend cover and Altman z-score.
30 April 2009
11
Overlay on a multifactor model

as a knock out signal
add value without
significantly
increasing turnover.
Next, we look at a different way to combine the Williams %R into a multifactor model. Given
the high turnover of the signal we look to test the Williams %R as a knockout filter for a
model. We use the same four factor model of equally weighted value, revisions, quality and
price momentum and again tested in the MSCI Australia universe. This time we divide the
standard model and the Williams %R technical indicator into three equal weighted groupings.
From there we knock out the stocks that appear to be overbought or oversold according to
the Williams %R indicator.
The annualised active mean returns of the standard model conditioned on the Williams %R
are show in Figure 17. The key corners of interest are the top right and bottom left. The top
right portfolio represents the stocks that score favourably on our standard model (long
portfolio) but look overbought according to the Williams %R. We would expect these stocks
to underperform. The bottom left portfolio represents stocks that are poor in our standard
portfolio but have been recently oversold. We would expect these to outperform.
Fig 17 Annualised active returns of filtered portfolios
Standard
Model
Williams %R
Over
sold
Middle
Over
bought
High
7.7
3.8
-2.4
3.3
Middle
5.8
1.5
-4.0
0.8
3.4
5.7
-4.5
0.3
-11.2
-6.0
-4.1
Low
Unconditional
Unconditional
Source: FactSet, Macquarie Research, April 2009
Further breaking down the above diagram (by corners):

For stocks with strong momentum that appear to be overbought (top right) we would
expect to see underperformance. This is indeed the case with the annualised market
relative return yielding -2.4% vs. 3.3% for the standard unconditional model (Figure 17).
Equally, we would expect strong performance from stocks with strong price momentum
that are oversold on technical signals (top left). This combination of characteristics yields
+7.7% (Figure 17).
Conversely, when looking at the portfolio that has poor momentum and appears
overbought on technical signals (highlighted bottom right) we would expect to see worse
performance. This is the case with this combination yielding -11.2% vs. -4.1% for the
unconditional portfolio (Figure 17).
Finally, we would expect better performance from the combination of weak momentum and
technically oversold stocks (bottom left). This again is true with the new combination
yielding +3.4% compared to -4.1%.
30 April 2009
12
Using the Williams %R signal as a knockout filter for a standard quant model appears to be
a practical way to add value to an investment process with technical indicators. The filter
helps to avoid stocks in the top tertile that have been overbought and those in the bottom
tertile that have been oversold. It is worth noting that the filtering methodology produces
more or less the same turnover as the standard model.
Figure 18 shows the cumulative returns of our filtered portfolio versus the standard unfiltered
portfolio. The overlay model has outperformed the standard model with Williams %R
consistently since 1994 and has done so with a lower turnover level of 101% 2-way per
month, vs. 148%. The overlay portfolio produced an increase in after cost information ratio
from 0.39 to 0.66 from the standard model with Williams %R.
Fig 18 Cumulative returns for enhanced portfolio

450
Fig 19 Monthly Q1-Q5 two way turnover
Standard Model (After Costs)
160%
400
Standard Model With Williams (After Costs)
140%
350
Standard Model With Williams Overlay (30bps Tcost)
120%
300
100%
250
Fig 20
Dec-08
Dec-07
Dec-06
Dec-05
Dec-04
Dec-03
Dec-02
Dec-01
Dec-00
Dec-99
Dec-98
0%
Dec-97
20%
Dec-96
40%
50
Dec-95
60%
100
Dec-94
80%
150
Dec-93
200
Standard Model
Standard Model With

Williams
Standard Model With

Williams Overlay
Annualised returns of model portfolios
50%
40%
Annual Return
30%
20%
10%
0%
-10%
-20%
-30%
Standard Model
Standard Model With Williams Overlay
30 April 2009
13
References
Andersen, T.G., Bollerslev, T., Answering the sceptics: yes, standard volatility models do
provide accurate Forecasts, International Economic Review Vol 39, (1998)
Bessembinder, H., Chan, K., Market efficiency and the returns to technical analysis.
Financial Management Vol 272 (1998).
Brock, W., Lakonishok, J., LeBaron, B., Simple technical trading rules and the stochastic
properties of stock returns Journal of Finance vol 485 (1992)
Davies, C. Quantamentals: Do Tehcnicals add value?, Macquarie Research Equities, (2009)
De Souza, R., Macquarie Alpha Model: The alpha model goes under the knife, Macquarie
Research Equities, (2009)
Fong, W., Yong, L. Chasing trends: recursive moving average trading rules and Internet
stocks Journal of Empirical Finance Vol 12 (2005)
Kavajecz, K.A., Odders-White, E.R., Technical analysis and liquidity provision Review of
Financial Studies Vol 17 (2005)
Kwon, K-Y., Kish, R.J., A comparative study of technical trading strategies and return
predictability: an extension of Brock, Lakonishok, and LeBaronk (1992) using NYSE and
NASDAQ indices Quarterly Review of Economics and Finance Vol 42 (2002)
Lee, D.D., Chan, H., Faff, R.W., Kalev, Short-term contrarian investing it is profitable
yes and no Journal of Multinational Financial Management Vol 13 (2003)
Lo, A.W., Mamaysky, H., Wang, J., Foundations of technical analysis: computational
algorithms, statistical inference, and empirical implementation Journal of Finance Vol 55
(2000)
Marshall, B.R., Cahan, R.H., Cahan, J.M. Does intraday technical analysis in the U.S. equity
market have value? Journal of Empirical Finance Vol 15 (2008)
Ready, M.J., Profits from technical trading rules Financial Management Vol 313 (2002).
Sullivan, R., Timmermann, A., White, H., Data-snooping, technical trading rule performance,
and the bootstrap Journal of Finance Vol 245 (1999)
30 April 2009
14
Appendices factor definitions

Fig 21 Technical definitions
Relative Strength Index
Developed by J. Welles Wilder in 1978 (New Concepts in Technical Trading Systems), it measures price relative to past prices. The indicator
compares the magnitude of a stocks recent gains to its recent losses and in effect measures the momentum behind recent price movements. That is
it measures the internal price strength of a security.
RSI = 100 100/(1+RS) RS = abs((Total gains over n periods/n) / (Total losses over n periods/n))
When the RSI falls below 30, it is deemed as a buy and when it rises above 70 it is deemed as sell. However, many practitioners will wait for the
signal to turn. In terms of testing the signal given a high score is deemed as overbought we multiply the RSI by -1 so that overbought stocks score
lowly. Additionally, given that a stock may continuously trade well within the bounds of the standard cut offs, we also look at the current level of the
RSI of a stock relative to its own long term average and variance that is we normalise the RSI for each individual stock we test on a rolling window
of 3, 6 and 12 months.
Williams %R
Developed by Larry Williams, it measures the relative location of the closing price in relation to the high-low range over a set time period with the
intention to identify over extended moves in a non-trending market. The indicator will typically peak and turn down a few days before the price peak
of the underlying security and vice versa when forming a trough. It is calculated by first determining the highest high and lowest low over the
preceding n periods, then taking the ratio of the highest high minus the current price divided by the highest high minus the lowest low. Values range
between -100 and 0.
W = -100 * ((Hn Tc)/(Hn-Ln)) Hn = highest price in n periods; Ln = lowest price in n periods, Tc = todays Close
Values between 0 & -20 it is deemed as overbought; between -80 and -100 it is deemed as oversold. Like the RSI we once again multiply the score
by -1 to make sure high scoring (overbought stocks close to 0) rank lowly. Additionally, we also test a normalised version of the signal on a 3-, 6- and
12-month basis.
MACD - Moving Average Convergence/Divergence
Developed by Gerald Appel in the 1960s, the MACD (Moving Average Convergence/Divergence) shows changes in relative movement of shorter and
longer moving averages based upon a price series. The indicator has both trending and oscillator characteristics and tends to give less false signals
than the faster RSI and Stochastic Oscillator.
The MACD Line is the difference between the longer exponentially weighted moving average (typically 26 days) and the shorter exponentially
weighted moving average (typically 12 days). A trigger line is then 9-day exponentially weighted moving average of the MACD. In 1986, Thomas
Aspray added the MACD histogram which is simply a graphical representation of the difference between the MACD and trigger line. A buy signal is
deemed to arise when the MACD rises through and above the trigger line, and the opposite is true for a sell signal.
Given that were measuring this as a style factor, we take the view the more positive the difference (or the more positive the histogram) the more
overbought a stock. Consequently, we multiply the MACD divergence by -1 so that a large divergence scores badly. We also test this factor on a
normalised basis relative to a stocks own history over 3-, 6- and 12-month periods.
Stochastic Oscillator
Developed by George Lane in the 1950s, this metric is a momentum-based price oscillator which aims to catch tops and bottoms early. When prices
close near to their highs, it is an indication of strength (bull) and vice versa when near lows (bear). In practice two oscillators are calculated. These
are more commonly referred to the fast (%K) and slow (%D). The %K is the same as the Williams %R but ranges from 0 to 100 using the formula
below:
%K = ((Tc Ln)/(Hn-Ln))*100 Hn = highest price in n periods; Ln = lowest price in n periods, Tc = todays Close
The %D is a simple moving average of the %K using an n day window
For the purposes of our research we create a ratio of the %K relative to %D. When this ratio is positive then this is deemed as attractive and vice
versa. Like the other metrics we also normalise the signal relative to a stocks own history and using a rolling 3-, 6- and 12-month window.
30 April 2009
15
Fig 21 Technical definitions (continued)

Developed by Donald Lambert initially for use on commodities, it now is frequently used on equities and currencies. The CCI is calculated as the
difference between the typical price and its simple moving average, divided by the mean of the typical price. The index is usually scaled by a factor of
1/0.015 to provide more readable numbers.
CCI = (1/0.015) * (Typical Price Simple Moving Average of Typical Price)/Mean Deviation
Typical Price = average of high, low and close price
Mean Deviation = Average(Abs(SMA Typical Price today Typical Price for past n periods))
It is used to identify divergences and therefore acts as an overbought/oversold indicator. It typically oscillates around 0 ranging between +/- 100.
When it rises above 100 it implies and overbought position and when it falls below -100 it represents an oversold position.
Money Flow Index
Similar to On Balance Volume, the Money Flow Index also incorporates the momentum characteristics based upon the relative location of the price. It
is interpreted in a similar manner to the RSI in that its value ranges between 0 and 100 and a value < 20 is oversold while a value > 80 is overbought.
It is calculated as follows
Typical Price = (High Price + Low Price + Close Price)/3
Money Flow = Typical Price * Volume
Money Ratio = Positive Money Flow/Negative Money Flow where positive money flow is when todays typical price > yesterdays typical price. The
Positive Money Flow is then the sum of that series over the past n periods. Negative money flow is when todays typical price < yesterdays typical
price.
The Money Flow Index = 100 (100/(1+Money Ratio))
Like the RSI we multiply this signal by * -1 so that a high score (>80 overbought) actually ranks lowly. Additionally, we also test a normalised version
of the signal over a 3-, 6- and 12-month basis.
Chaikin Money Flow
Developed by Marc Chaikin, this indicator is an enhancement of the Accumulation / Distribution Index. It simply takes the ADI and then scales this by
the total volume traded over n periods (traders typically use 21 days).
In terms of usage, the basic rule of thumb is that is the value is > 0 then this is deemed as bullish and < 0 is bearish. If the value is more that +/- 0.25
this is viewed as being a strong trend. A divergence can show up when the CMF makes a new high but the prices action creates a new flow. This
implies that there is less selling pressure pushing the stock lower and therefore a bounce could occur.
On Balance Volume
Popularised by J Granville in his 1963 book Granville's New Key to Stock Market Profits, the metric was first investigated by Woods & Vignolia in
1940. The indicator relates volume to price change and indicates if volume is flowing into or out of a security.
It adds and subtracts volume to a running total depending on whether price moves up or down.
It is often used as an indicator of stocks under accumulation for a possible take over. The rationale for this is that potential bidders/acquirers of large
stock positions tend to place bids just below the market and only take the ask price when a sizeable block is offered. Consequently, the indicator
moves higher during a time when the stock price remains relatively flat.
In terms of analysis investors should be wary of simply looking at the raw level and actually the direction tends to be more important as it indicated
the flow. The metric is viewed as leading indicator of price trends and consequently a divergence between the indicator and price could signal that a
reversal may about to occur. Having looked at the strategy it appears that like other metrics above it actually acts as a contrarian indicator in that
when the OBV reaches new highs there are likely to be market participants taking profits. Consequently when we look at the signal and normalise it
relative to a stocks own history we multiply the factor by -1.
Force Index
Developed by Alexander Elder for intra day data this indicator measures the bullish (bearish) force during each upward (downward) movement - the
force of each movement is defined by its trend, range and volume. While the index can be used on its own, practitioners typically prefer to use a
moving average. A short moving average is used to identify when to open and close positions, while a longer moving average is used to show trends
and their changes.
We have adapted the calculation to use daily data as opposed to tick data.
Force = (Price Today Price Yesterday) * Volume
Given the above, If the close of the current is > close of previous then force is positive and if the close of the current is < close of previous then force
is negative. Additionally, the large the value can be either be driven by the greater difference in price or a greater the transaction volume. Like all
oscillators, this indicator indicates a buy when it rises above the centre line and trend lines can be applied to determine the strength of the
movement. Like the other oscillators, we multiply this factor by -1 so that a high score scores badly.
30 April 2009
16
Fig 21 Technical definitions (continued)

Developed by Marc Chaikin, this indicator is a cumulative total volume indicator which uses both prices and volume. The volume added each day is a
function of the relative strength of price within the day. The nearer the closing price to its high of the day, the higher the added share of volume will
be added and vice versa. This weighting coefficient is calculated as:
((Close Low) (High Close)) / (High Low) - This therefore ranges between +/-1.
Consequently the incremental value added to the index on any given day is as follows:
AD = (((Close Low) (High Close)) / (High Low)) * Volume
When this indicator grows, it indicates that there is accumulation of a particular security as the majority of the share of the share of the volume is
related to an upward trend in prices.
Average Directional Movement Index (ADX)
Developed by Wilder, the ADX attempts to evaluate the strength of a current trend, be it up or down. Although market direction is integral to the
calculation of the ADX, it itself is not a directional indicator and therefore should be used in conjunction with other indicators.
The ADX is derived from two other indicators the positive directional indicator (+DI) and negative directional indicator (-DI). When the +DI rises above
the DI, then a buy signal is deemed.
The ADI is then the ratio of +DI/-DI on a moving average basis. The scale of the ADX ranges from 0 to 100 with low values (<20) indicating a weak
trend and high (>40) a strong trend. When the signal moves up and through 20, it indicates that a trading range is potentially ending and a trend is
forming. When declining from above 40, this can indicate that a trend could be slowing and a trading range may be developing.
Average True Range
Developed by J. Welles Wilder, this metric dynamically measures the typical range between prices (high, low, close) over a set period. The True
range is the greater of the following three measures:
Todays High Todays Low
Absolute( Todays High Yesterdays Close)
Absolute( Todays Low Yesterdays Close)
The Average True Range is the moving average of the true range values.
The basic interpretation is the greater the value, the greater the possibility for a trend reversal; the smaller the value, the weaker the trend.
Bollinger Bands
Developed by John Bollinger, these bands are used to visually judge the relationship between price and volatility changes.
Bollinger bands consist of a centre line and an upper and lower band around that centre line. The centre line is a moving average, which could be
simple or exponential, and we use a simple moving average of 20 days which is in line with providers such as FactSet and Bloomberg. The bands
are then formed by calculating two standard deviations over the same period which are then added and subtracted from the centre line.
They differ from simple bands around a mean in that in trending markets the bands will tend to expand, while in range-bound markets they will
contract. The theory is that in range bound markets resistance should be found at the upper and lower levels. However, should the price break
through, then this could indicate a breakout.
Generally, Bollinger Bands can be useful to identify buy and sell signals but are not in themselves designed to determine the future direction of a
security. They do, however, help identify periods of high and low volatility (bandwidth) and when prices are extreme and potentially unsustainable.
30 April 2009
17
Technical factor performance

Williams %R - Time Series Normalised
Last 5
Years
5.4%
7.9%
73.3%
0.68
5.29
All History
Average IC
Std Deviation IC
Success Rate
Avg IC / Std Dev IC
t-stat
5.6%
9.6%
73.7%
0.59
7.68
Last 3
Last 1 Year
Years
4.8%
5.9%
8.7%
12.3%
63.9%
50.0%
0.55
0.48
3.29
1.65
40%
30%
20%
10%
0%
-10%
-20%
-30%
Monthly IC
08
D
ec
D
ec
07
06
05
D
ec
D
ec
D
ec
04
03
02
D
ec
D
ec
01
00
D
ec
D
ec
D
ec
99
98
97
D
ec
96
D
ec
D
ec
D
ec
D
ec
94
95
-40%
12M Average IC
IC Horizon
Average IC
5.6%
1.4%
-0.3%
-1.8%
-1.5%
1 month
3 month
6 month
12 month
24 month
8.0%
6.0%
4.0%
2.0%
0.0%
-2.0%
-4.0%
1
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
IC Decay
Accumulated IC
Long Short Q1-Q5

Last 5
Last 3
Last Year
Years
Years
Quintile 1 Excess Return *
4.4%
3.8%
3.5%
3.7%
-8.3%
-10.4%
-11.9%
-23.3%
13.2%
15.1%
16.4%
32.1%
10.2%
10.1%
12.2%
19.6%
1.29
1.50
1.34
1.64
1913%
1934%
1922%
1934%
10
Excess Market Return
All History
Long Short Q1-Q5 Rolling Excess Returns
08
07
D
ec
D
ec
D
ec
06
05
04
D
ec
D
ec
Bottom Quintile
13.1%
10.2%
1.28
1913%
Last 3
Last 1 Year
Years
16.4%
32.1%
12.2%
19.6%
1.34
1.64
1922%
1934%
Long Short Returns
50%
Last 5
Years
15.1%
10.1%
1.50
1934%
All History
Q1 - Q5 Return (p.a.)
Q1 - Q5 Volatility (p.a.)
Q1 - Q5 Information Ratio
Q1 - Q5 Turnover (p.a. 1 way)
03
02
Top Quintile
D
ec
D
ec
D
ec
01
00
99
D
ec
D
ec
D
ec
98
97
96
D
ec
D
ec
D
ec
D
ec
95
94
40%
30%
20%
10%
0%
Long Short Q1-Q5 Turnover 1-way
08
D
ec
07
D
ec
D
ec
06
05
D
ec
D
ec
04
03
D
ec
2500%
Last 5 Years Last 3 Years Last 1 Year
1913%
1934%
1922%
1934%
2000%
Turnover
All History
Average Annual L/S Turnover
02
01
LS Return
D
ec
00
D
ec
D
ec
D
ec
99
98
97
D
ec
D
ec
D
ec
96
95
D
ec
D
ec
94
-10%
1500%
1000%
500%
08
ec
D
ec
D
07
06
ec
05
04
ec
03
D
Turnover
ec
ec
02
01
ec
00
D
ec
99
ec
98
ec
97
ec
D
ec
96
ec
D
95
ec
D
ec
94
0%
12M Rolling Turnover
Pure Factor Return
Average pure factor return (p.a.)

Pure Factor Volatility
Pure Factor Sharpe
5.11%
2.82%
1.81
Last 5
Years
5.11%
2.88%
1.78
Last 3
Last 1 Year
Years
6.23%
9.68%
3.44%
5.20%
1.81
1.86
12%
10%
Factor Return
All History
8%
6%
4%
2%
0%
-2%
07
06
08
D
ec
D
ec
04
03
05
D
ec
D
ec
D
ec
01
02
D
ec
D
ec
00
Factor Return
D
ec
98
97
96
99
D
ec
D
ec
D
ec
D
ec
D
ec
95
D
ec
D
ec
94
-4%
Annualised Factor Return
Source: IBES, MSCI, and Macquarie Research, April 2009
30 April 2009
18
Williams %R - Time Series Normalised - continued

Fractiles
Excess
Return
3.8%
1.7%
1.7%
-0.3%
-5.8%
Volatility
17.0%
16.1%
16.2%
15.0%
15.8%
Informatio
n Ratio
0.22
0.11
0.11
-0.02
-0.36
100
Cumulative Return
Q1
Q2
Q3
Q4
Q5
Market
Annual
Return
14.3%
12.3%
12.2%
10.2%
4.7%
10.5%
10
All History
Q1
Q2
Q3
Q4
Q5
53.4%
50.2%
50.6%
48.6%
45.9%
Last 5
Years
52.5%
51.3%
51.5%
48.5%
45.7%
Last 3
Last 1 Year
Years
52.7%
52.0%
51.0%
51.6%
50.9%
53.1%
48.8%
50.5%
46.0%
42.2%
% Cross sectional hit rate
Fractile Cross sectional Hit Rate
Q2
Q3
Q4
Q5
08
D
ec
D
ec
07
06
05
D
ec
D
ec
D
ec
04
03
02
D
ec
D
ec
D
ec
D
ec
01
00
99
98
D
ec
Q1
D
ec
97
96
D
ec
95
D
ec
D
ec
D
ec
94
Market
60.0%
58.0%
56.0%
54.0%
52.0%
50.0%
48.0%
46.0%
44.0%
42.0%
40.0%
Q1
Q2
All History
Q3
Last 5 Years
Q4
Last 3 Years
Q5
Last 1 Year
Size
Large
Small
All
Std Dev IC
Small
Large
All
All History
6.15%
5.13%
5.63%
All History
13.06%
12.09%
9.61%
Last 5
Years
5.68%
5.30%
5.42%
Latest No. of
Last 3
Last 1 Year
Stocks
Years
94
6.40%
7.35%
191
4.07%
5.37%
285
4.76%
5.87%
Last 5
Years
12.85%
8.99%
7.93%
Last 3
Last 1 Year
Years
12.88%
14.90%
9.09%
12.91%
8.68%
12.31%
Last 5
Years
Last 3
Years
8.0%
Average IC
7.0%
6.0%
5.0%
4.0%
3.0%
2.0%
1.0%
0.0%
Large
All History
Small
Last 3 Years
Last 5 Years
All
Last 1 Year
All History
Seasonality
Samples
Utilities
Telecomm.
Services
Information
Tech
Financials
Health Care
2%
D
ec
N
ov
08
D
ec
07
D
ec
05
04
03
06
D
ec
D
ec
D
ec
02
D
ec
Stocks with Factor

% Coverage by Market Cap
D
ec
01
D
ec
00
D
ec
99
D
ec
98
D
ec
97
D
ec
96
% of Market Covered
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
D
ec
D
ec
em
be
r
em
be
r
r
ct
o
be
Se
p
gu
Au
te
m
be
st
y
Ju
l
ne
Ju
M
ay
Ap
ril
ar
ch
ar
y
ru
400
350
300
250
200
150
100
50
0
94
100%
Fe
b
nu
ar
Last 3
Last 1 Year
Years
322
319
99%
4%
95
99%
6%
D
ec
99%
8%
0%
Number of Stocks
228
Last 5
Years
291
All History
Coverage of available stocks (%)
Last 3 Years
10%
15
15
15
14
14
14
14
14
14
14
14
14
Factor Coverage
Average number of stocks
Last 5 Years
12%
Information
IC Std Dev
Coefficient
10.9%
9.1%
1.2%
11.2%
1.9%
9.2%
7.4%
5.4%
6.9%
11.5%
8.8%
9.3%
9.8%
11.1%
2.6%
11.7%
3.7%
8.7%
7.7%
7.8%
2.9%
7.0%
3.9%
8.3%
Ja
January
February
March
April
May
June
July
August
September
October
November
December
Consumer
Staples
8.92%
-0.58%
7.85%
7.43%
4.13%
8.85%
6.82%
15.69%
2.50%
14.37%
Consumer
Discretionary
4.64%
1.36%
8.30%
5.77%
3.27%
8.94%
6.21%
15.93%
9.83%
11.56%
Industrials
3.06%
5.97%
5.93%
5.44%
7.91%
4.32%
8.39%
7.49%
13.53%
9.99%
18%
16%
14%
12%
10%
8%
6%
4%
2%
0%
-2%
Materials
Energy
Materials
Industrials
Consumer Discretionary
Consumer Staples
Health Care
Financials
Information Tech
Telecomm. Services
Utilities
Latest
Number of
Stocks
32
57
45
31
16
10
73
3
3
15
Energy
All History
Sector
% Coverage of Universe
30 April 2009
19
Force Index - Time Series Normalised

Last 5
Years
4.6%
7.3%
73.3%
0.64
4.93
All History
Average IC
Std Deviation IC
Success Rate
Avg IC / Std Dev IC
t-stat
4.4%
8.7%
69.0%
0.51
6.74
Last 3
Last 1 Year
Years
3.7%
3.5%
7.2%
9.1%
69.4%
58.3%
0.52
0.38
3.11
1.32
40%
30%
20%
10%
0%
-10%
-20%
-30%
08
07
D
ec
D
ec
D
ec
06
05
04
D
ec
D
ec
D
ec
03
02
01
Monthly IC
D
ec
D
ec
D
ec
00
99
98
D
ec
D
ec
97
96
D
ec
D
ec
D
ec
D
ec
95
94
-40%
12M Average IC
IC Horizon
Average IC
4.4%
2.2%
1.0%
-0.1%
-0.3%
1 month
3 month
6 month
12 month
24 month
6.0%
4.0%
2.0%
0.0%
-2.0%
-4.0%
1
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
IC Decay
Accumulated IC
Long Short Q1-Q5

Last 5
Last 3
Last Year
Years
Years
6.3%
6.3%
7.3%
15.6%
-3.2%
-5.3%
-4.0%
-2.2%
9.4%
12.0%
11.4%
17.7%
8.8%
7.1%
8.0%
11.7%
1.07
1.68
1.44
1.52
1855%
1899%
1894%
1858%
10
All History
LS Return
08
ec
07
06
ec
D
ec
ec
05
04
ec
D
D
ec
08
07
06
D
ec
D
ec
D
ec
05
04
D
ec
03
D
ec
02
2500%
1855%
1899%
1894%
1858%
2000%
Turnover
All History
D
ec
00
01
D
ec
99
D
ec
98
D
ec
D
ec
97
96
D
ec
D
ec
D
ec
D
ec
94
9.4%
8.8%
1.06
1855%
Bottom Quintile
35%
30%
25%
20%
15%
10%
5%
0%
-5%
-10%
-15%
95
All History
Last 3
Last 1 Year
Years
11.4%
17.7%
8.0%
11.7%
1.44
1.52
1894%
1858%
Long Short Returns
Last 5
Years
12.0%
7.1%
1.68
1899%
03
02
D
Top Quintile
ec
ec
D
ec
01
00
99
D
ec
ec
98
97
ec
96
ec
D
ec
D
ec
ec
94
95
1500%
1000%
500%
Turnover
08
D
ec
07
D
ec
06
D
ec
05
D
ec
04
03
D
ec
02
D
ec
D
ec
00
01
D
ec
99
D
ec
98
D
ec
D
ec
97
96
D
ec
D
ec
D
ec
D
ec
94
95
0%
Pure Factor Return

Pure Factor Sharpe
4.17%
2.50%
1.67
Last 5
Years
4.34%
2.34%
1.86
Last 3
Last 1 Year
Years
4.71%
6.93%
2.57%
3.51%
1.83
1.97
12%
10%
Factor Return
All History
8%
6%
4%
2%
0%
-2%
07
06
08
D
ec
D
ec
04
03
05
D
ec
D
ec
D
ec
01
02
D
ec
D
ec
00
Factor Return
D
ec
98
97
96
99
D
ec
D
ec
D
ec
D
ec
D
ec
95
D
ec
D
ec
94
-4%
30 April 2009
20
Force Index - Time Series Normalised - continued

Fractiles
Excess
Return
5.8%
1.6%
-1.3%
-3.0%
-2.4%
Volatility
15.7%
16.9%
15.3%
16.6%
15.5%
Informatio
n Ratio
0.37
0.10
-0.09
-0.18
-0.16
100
Cumulative Return
Q1
Q2
Q3
Q4
Q5
Market
Annual
Return
16.3%
12.2%
9.2%
7.5%
8.1%
10.5%
10
All History
Q1
Q2
Q3
Q4
Q5
53.6%
50.0%
49.1%
47.9%
48.0%
Last 5
Years
52.8%
51.0%
48.9%
48.1%
48.6%
Last 3
Last 1 Year
Years
52.7%
54.1%
50.8%
48.0%
48.5%
50.4%
48.2%
46.5%
49.3%
50.6%
Q2
Q3
Q4
Q5
08
07
D
ec
D
ec
D
ec
06
05
04
D
ec
D
ec
D
ec
D
ec
03
02
01
00
D
ec
D
ec
99
98
D
ec
Q1
D
ec
97
96
D
ec
95
D
ec
D
ec
D
ec
94
Market
60.0%
58.0%
56.0%
54.0%
52.0%
50.0%
48.0%
46.0%
44.0%
42.0%
40.0%
Q1
Q2
All History
Q3
Last 5 Years
Q4
Last 3 Years
Q5
Last 1 Year
Size
Large
Small
All
Std Dev IC
Small
Large
All
All History
4.03%
4.60%
4.45%
All History
12.16%
10.73%
8.66%
Last 5
Years
2.86%
5.23%
4.62%
Latest No. of
Last 3
Last 1 Year
Stocks
Years
94
2.04%
0.93%
188
4.22%
4.60%
282
3.71%
3.46%
Last 5
Years
12.38%
7.89%
7.25%
Last 3
Last 1 Year
Years
13.76%
14.48%
6.71%
8.74%
7.16%
9.09%
Last 5
Years
Last 3
Years
6.0%
Average IC
5.0%
4.0%
3.0%
2.0%
1.0%
0.0%
Large
All History
Small
Last 3 Years
Last 5 Years
All
Last 1 Year
20%
15%
10%
5%
0%
All History
Seasonality
Utilities
Telecomm.
Services
Information
Tech
Health Care
7%
15
15
15
14
14
14
14
14
14
14
14
14
6%
5%
4%
3%
2%
1%
D
ec
N
ov
08
D
ec
07
D
ec
05
04
03
06
D
ec
D
ec
D
ec
02
Stocks with Factor

D
ec
01
D
ec
D
ec
00
D
ec
99
D
ec
98
D
ec
97
D
ec
96
% of Market Covered
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
D
ec
D
ec
em
be
r
em
be
r
r
be
ct
o
O
be
Se
p
Au
te
m
gu
st
y
Ju
l
ne
ay
M
Ju
Ap
ril
ar
ch
ru
Fe
b
nu
ar
99%
94
98%
400
350
300
250
200
150
100
50
0
95
99%
Last 3
Last 1 Year
Years
320
316
D
ec
99%
ar
y
0%
Number of Stocks
228
Last 5
Years
290
All History
Last 3 Years
8%
Samples
Factor Coverage
Last 5 Years
9%
Information
IC Std Dev
Coefficient
0.9%
6.2%
0.7%
9.7%
3.5%
7.4%
4.3%
10.3%
8.1%
8.4%
6.9%
9.8%
8.4%
9.5%
5.6%
7.1%
1.0%
8.4%
8.4%
9.7%
2.7%
8.1%
3.4%
5.6%
Ja
January
February
March
April
May
June
July
August
September
October
November
December
Financials
-5%
Consumer
Staples
4.41%
0.50%
2.40%
7.12%
-1.89%
8.97%
5.65%
10.40%
10.28%
13.60%
Consumer
Discretionary
3.89%
2.47%
3.63%
5.84%
-0.23%
10.66%
5.38%
19.09%
2.00%
11.79%
Industrials
3.40%
5.25%
6.35%
4.78%
5.36%
8.36%
6.01%
8.56%
3.58%
1.00%
25%
Materials
Energy
Materials
Industrials
Consumer Staples
Health Care
Financials
Information Tech
Telecomm. Services
Utilities
Latest
Number of
Stocks
32
56
45
31
16
10
71
3
3
15
Energy
All History
Sector
30 April 2009
21
Bollinger Band - Time Series Normalised

Last 5
Years
4.6%
9.0%
71.7%
0.51
3.99
All History
Average IC
Std Deviation IC
Success Rate
Avg IC / Std Dev IC
t-stat
4.4%
10.1%
67.3%
0.44
5.77
Last 3
Last 1 Year
Years
4.2%
4.6%
9.7%
12.7%
69.4%
58.3%
0.43
0.36
2.59
1.26
40%
30%
20%
10%
0%
-10%
-20%
-30%
Monthly IC
08
07
D
ec
06
D
ec
05
D
ec
D
ec
D
ec
04
03
02
D
ec
D
ec
01
00
D
ec
D
ec
99
98
D
ec
D
ec
97
96
D
ec
D
ec
D
ec
D
ec
95
94
-40%
12M Average IC
IC Horizon
Average IC
4.4%
0.6%
-0.5%
-2.3%
-1.8%
1 month
3 month
6 month
12 month
24 month
6.0%
4.0%
2.0%
0.0%
-2.0%
-4.0%
1
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
IC Decay
Accumulated IC
Long Short Q1-Q5

10
Last 5
Last 3
Last Year
All History
Years
Years
3.1%
2.5%
1.4%
4.0%
-4.8%
-6.2%
-8.0%
-14.0%
7.7%
8.6%
9.3%
18.3%
10.3%
10.5%
12.5%
20.3%
0.76
0.82
0.75
0.90
1939%
1955%
1969%
1977%
08
D
ec
07
D
ec
D
ec
06
05
04
D
ec
D
ec
D
ec
08
07
D
ec
06
D
ec
D
ec
05
04
D
ec
03
02
2500%
1939%
1955%
1969%
1977%
2000%
Turnover
All History
D
ec
01
00
LS Return
D
ec
D
ec
99
D
ec
D
ec
98
97
D
ec
D
ec
96
95
D
ec
D
ec
D
ec
Last 3
Last 1 Year
Years
9.3%
18.3%
12.5%
20.3%
0.75
0.90
1969%
1977%
Bottom Quintile
35%
30%
25%
20%
15%
10%
5%
0%
-5%
-10%
94
7.7%
10.3%
0.75
1939%
Long Short Returns
Last 5
Years
8.6%
10.5%
0.82
1955%
All History
03
02
01
00
Top Quintile
D
ec
D
ec
D
ec
99
D
ec
98
D
ec
D
ec
97
96
D
ec
D
ec
D
ec
D
ec
94
95
1500%
1000%
500%
D
ec
08
D
ec
07
06
D
ec
05
D
ec
04
D
ec
03
02
D
ec
01
00
Turnover
D
ec
D
ec
99
D
ec
98
D
ec
D
ec
D
ec
96
D
ec
95
D
ec
94
D
ec
97
0%
Pure Factor Return
3.42%
2.83%
1.21
Last 5
Years
3.51%
3.06%
1.15
Last 3
Last 1 Year
Years
4.75%
7.19%
3.52%
5.45%
1.35
1.32
10%
8%
Factor Return
All History
Pure Factor Sharpe
6%
4%
2%
0%
-2%
07
06
08
D
ec
D
ec
04
03
05
D
ec
D
ec
D
ec
01
02
D
ec
D
ec
00
Factor Return
D
ec
98
97
96
99
D
ec
D
ec
D
ec
D
ec
D
ec
95
D
ec
D
ec
94
-4%
30 April 2009
22
Bollinger Band - Time Series Normalised - continued

Fractiles
Excess
Return
2.7%
1.5%
0.5%
-1.1%
-3.5%
Volatility
16.7%
16.0%
16.3%
16.0%
14.9%
Informatio
n Ratio
0.16
0.09
0.03
-0.07
-0.24
100
Cumulative Return
Q1
Q2
Q3
Q4
Q5
Market
Annual
Return
13.2%
12.1%
11.1%
9.5%
7.0%
10.6%
10
All History
Q1
Q2
Q3
Q4
Q5
53.3%
50.0%
49.5%
49.7%
46.5%
Last 5
Years
52.1%
51.6%
51.0%
48.3%
46.5%
Last 3
Last 1 Year
Years
51.4%
50.9%
51.6%
51.8%
51.6%
52.3%
49.4%
50.7%
45.7%
43.8%
Q2
Q5
08
07
06
D
ec
D
ec
D
ec
D
ec
05
04
03
Q4
D
ec
D
ec
D
ec
D
ec
Q3
02
01
00
99
D
ec
98
D
ec
Q1
D
ec
97
96
D
ec
95
D
ec
D
ec
D
ec
94
Market
60.0%
58.0%
56.0%
54.0%
52.0%
50.0%
48.0%
46.0%
44.0%
42.0%
40.0%
Q1
Q2
All History
Q3
Last 5 Years
Q4
Last 3 Years
Q5
Last 1 Year
Size
Large
Small
All
Std Dev IC
Small
Large
All
All History
5.81%
3.40%
4.44%
All History
13.49%
11.72%
10.07%
Last 5
Years
6.35%
3.65%
4.65%
Latest No. of
Last 3
Last 1 Year
Stocks
Years
94
6.96%
10.10%
191
2.97%
2.74%
285
4.17%
4.64%
Last 5
Years
14.36%
9.70%
9.04%
Last 3
Last 1 Year
Years
14.73%
15.24%
9.41%
12.89%
9.66%
12.72%
Last 5
Years
Last 3
Years
12.0%
Average IC
10.0%
8.0%
6.0%
4.0%
2.0%
0.0%
Large
Last 5 Years
All History
Small
Last 3 Years
All
Last 1 Year
All History
Seasonality
Samples
15
15
15
14
14
14
14
14
14
14
14
14
Utilities
Telecomm.
Services
Information
Tech
Financials
Health Care
4%
2%
0%
D
ec
N
ov
08
D
ec
07
D
ec
05
04
03
06
D
ec
D
ec
D
ec
02
Stocks with Factor

D
ec
01
D
ec
D
ec
00
D
ec
99
D
ec
98
D
ec
97
D
ec
96
% of Market Covered
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
D
ec
D
ec
em
be
r
em
be
r
r
ct
o
be
Se
p
gu
Au
te
m
be
st
y
Ju
l
ne
Ju
M
ay
Ap
ril
ar
ch
ar
y
ru
Fe
b
nu
ar
100%
400
350
300
250
200
150
100
50
0
94
99%
6%
95
99%
8%
D
ec
99%
Last 3
Last 1 Year
Years
322
319
Number of Stocks
229
Last 5
Years
291
All History
Last 3 Years
-2%
Factor Coverage
Last 5 Years
10%
Information
IC Std Dev
Coefficient
9.4%
11.9%
-0.2%
12.1%
1.6%
9.3%
5.4%
6.9%
6.4%
10.9%
7.1%
11.2%
9.0%
11.3%
1.6%
10.2%
1.4%
8.7%
7.7%
8.3%
1.8%
6.7%
2.2%
8.5%
Ja
January
February
March
April
May
June
July
August
September
October
November
December
Consumer
Staples
7.10%
2.23%
6.40%
4.68%
4.74%
9.36%
6.63%
18.67%
7.22%
13.20%
Consumer
Discretionary
3.50%
2.36%
6.53%
3.76%
2.96%
10.04%
5.79%
13.33%
2.67%
11.60%
Industrials
2.89%
5.07%
5.45%
5.28%
7.45%
2.49%
8.13%
2.73%
4.62%
5.55%
20%
18%
16%
14%
12%
10%
8%
6%
4%
2%
0%
Materials
Energy
Materials
Industrials
Consumer Staples
Health Care
Financials
Information Tech
Telecomm. Services
Utilities
Latest
Number of
Stocks
32
57
45
31
16
10
73
3
3
15
Energy
All History
Sector
30 April 2009
23
Commodity Channel Index - Time Series Normalised

Last 5
Years
4.0%
8.0%
66.7%
0.50
3.84
All History
Average IC
Std Deviation IC
Success Rate
Avg IC / Std Dev IC
t-stat
4.1%
9.1%
65.5%
0.45
5.84
Last 3
Last 1 Year
Years
3.7%
3.9%
8.5%
12.6%
63.9%
50.0%
0.44
0.31
2.64
1.08
40%
30%
20%
10%
0%
-10%
-20%
-30%
Monthly IC
08
07
D
ec
06
D
ec
05
D
ec
D
ec
D
ec
04
03
02
D
ec
D
ec
01
00
D
ec
D
ec
99
98
D
ec
D
ec
97
96
D
ec
D
ec
D
ec
D
ec
95
94
-40%
12M Average IC
IC Horizon
Average IC
4.1%
0.8%
-0.2%
-1.5%
-1.5%
1 month
3 month
6 month
12 month
24 month
5.0%
4.0%
3.0%
2.0%
1.0%
0.0%
-1.0%
-2.0%
-3.0%
1
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
IC Decay
Accumulated IC
Long Short Q1-Q5

10
Last 5
Last 3
Last Year
All History
Years
Years
4.1%
3.0%
4.1%
10.3%
-4.0%
-6.0%
-6.9%
-13.2%
8.0%
9.1%
11.0%
24.7%
9.1%
9.3%
11.3%
18.3%
0.88
0.98
0.98
1.35
1928%
1931%
1928%
1951%
08
D
ec
07
D
ec
D
ec
06
05
04
D
ec
D
ec
D
ec
08
07
D
ec
06
D
ec
D
ec
05
04
D
ec
03
02
2500%
1928%
1931%
1928%
1951%
2000%
Turnover
All History
D
ec
01
00
LS Return
D
ec
D
ec
99
D
ec
D
ec
98
97
D
ec
D
ec
96
95
D
ec
D
ec
D
ec
Last 3
Last 1 Year
Years
11.0%
24.7%
11.3%
18.3%
0.98
1.35
1928%
1951%
Bottom Quintile
30%
25%
20%
15%
10%
5%
0%
-5%
-10%
-15%
94
7.9%
9.1%
0.87
1928%
Long Short Returns
Last 5
Years
9.1%
9.3%
0.98
1931%
All History
03
02
01
00
Top Quintile
D
ec
D
ec
D
ec
99
D
ec
98
D
ec
D
ec
97
96
D
ec
D
ec
D
ec
D
ec
94
95
1500%
1000%
500%
D
ec
08
D
ec
07
06
D
ec
05
D
ec
04
D
ec
03
02
D
ec
01
00
Turnover
D
ec
D
ec
99
D
ec
98
D
ec
D
ec
D
ec
96
D
ec
95
D
ec
94
D
ec
97
0%
Pure Factor Return
3.11%
2.73%
1.14
Last 5
Years
3.05%
3.07%
0.99
Last 3
Last 1 Year
Years
4.33%
7.46%
3.61%
5.70%
1.20
1.31
10%
8%
Factor Return
All History
Pure Factor Sharpe
6%
4%
2%
0%
-2%
07
06
08
D
ec
D
ec
04
03
05
D
ec
D
ec
D
ec
01
02
D
ec
D
ec
00
Factor Return
D
ec
98
97
96
99
D
ec
D
ec
D
ec
D
ec
D
ec
95
D
ec
D
ec
94
-4%
30 April 2009
24
Commodity Channel Index - Time Series Normalised - continued

Fractiles
Excess
Return
3.6%
0.5%
0.8%
-1.7%
-3.0%
Volatility
16.2%
16.5%
15.8%
16.0%
15.1%
Informatio
n Ratio
0.22
0.03
0.05
-0.11
-0.20
100
Cumulative Return
Q1
Q2
Q3
Q4
Q5
Market
Annual
Return
14.1%
11.0%
11.3%
8.8%
7.5%
10.5%
10
All History
Q1
Q2
Q3
Q4
Q5
53.0%
50.4%
49.1%
49.2%
47.2%
Last 5
Years
52.2%
52.3%
49.7%
48.9%
46.6%
Last 3
Last 1 Year
Years
53.0%
54.2%
51.5%
50.7%
50.0%
50.9%
48.6%
48.7%
46.6%
45.2%
Q2
Q5
08
07
06
D
ec
D
ec
D
ec
D
ec
05
04
03
Q4
D
ec
D
ec
D
ec
D
ec
Q3
02
01
00
99
D
ec
98
D
ec
Q1
D
ec
97
96
D
ec
95
D
ec
D
ec
D
ec
94
Market
60.0%
58.0%
56.0%
54.0%
52.0%
50.0%
48.0%
46.0%
44.0%
42.0%
40.0%
Q1
Q2
All History
Q3
Last 5 Years
Q4
Last 3 Years
Q5
Last 1 Year
Size
Large
Small
All
Std Dev IC
Small
Large
All
All History
5.20%
3.18%
4.06%
All History
12.59%
11.03%
9.12%
Last 5
Years
5.03%
3.66%
3.97%
Latest No. of
Last 3
Last 1 Year
Stocks
Years
93
6.29%
6.59%
191
2.90%
3.32%
284
3.75%
3.93%
Last 5
Years
13.18%
8.75%
8.01%
Last 3
Last 1 Year
Years
12.63%
13.04%
9.04%
13.93%
8.51%
12.59%
Last 5
Years
Last 3
Years
7.0%
Average IC
6.0%
5.0%
4.0%
3.0%
2.0%
1.0%
0.0%
Large
All History
Small
Last 3 Years
Last 5 Years
All
Last 1 Year
20%
15%
10%
5%
0%
-5%
All History
Seasonality
Samples
Utilities
Telecomm.
Services
Information
Tech
Health Care
2%
0%
D
ec
N
ov
08
D
ec
07
D
ec
05
04
03
06
D
ec
D
ec
D
ec
02
Stocks with Factor

D
ec
01
D
ec
D
ec
00
D
ec
99
D
ec
98
D
ec
97
D
ec
96
% of Market Covered
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
D
ec
D
ec
em
be
r
em
be
r
r
ct
o
be
Se
p
gu
Au
te
m
be
st
y
Ju
l
ne
Ju
M
ay
Ap
ril
ar
ch
ar
y
ru
400
350
300
250
200
150
100
50
0
94
100%
Fe
b
nu
ar
Last 3
Last 1 Year
Years
322
319
99%
4%
95
99%
6%
D
ec
99%
8%
-2%
Number of Stocks
228
Last 5
Years
291
All History
Last 3 Years
10%
15
15
15
14
14
14
14
14
14
14
14
14
Factor Coverage
Last 5 Years
12%
Information
IC Std Dev
Coefficient
11.2%
8.2%
-0.6%
8.6%
-0.8%
8.9%
4.6%
5.6%
7.2%
11.5%
6.7%
7.7%
4.5%
10.0%
2.1%
11.3%
2.3%
9.6%
6.5%
7.5%
2.9%
7.2%
2.4%
7.2%
Ja
January
February
March
April
May
June
July
August
September
October
November
December
Financials
-10%
Consumer
Staples
6.98%
-1.85%
8.63%
5.67%
4.68%
13.65%
4.97%
20.74%
-3.61%
12.69%
Consumer
Discretionary
2.29%
0.90%
9.19%
4.32%
1.43%
11.72%
4.11%
10.80%
-7.17%
10.75%
Industrials
1.59%
5.01%
4.65%
6.06%
6.04%
3.83%
5.97%
0.85%
2.23%
3.85%
25%
Materials
Energy
Materials
Industrials
Consumer Staples
Health Care
Financials
Information Tech
Telecomm. Services
Utilities
Latest
Number of
Stocks
32
56
45
31
16
10
73
3
3
15
Energy
All History
Sector
30 April 2009
25
Relative Strength Indicator - 14Days - Time Series Normalised

Last 5
Years
4.2%
9.2%
70.0%
0.45
3.52
All History
Average IC
Std Deviation IC
Success Rate
Avg IC / Std Dev IC
t-stat
3.6%
10.4%
63.7%
0.35
4.59
Last 3
Last 1 Year
Years
3.2%
3.6%
9.5%
12.3%
66.7%
66.7%
0.34
0.29
2.03
1.01
40%
30%
20%
10%
0%
-10%
-20%
-30%
08
07
D
ec
ec
ec
06
05
04
ec
D
ec
ec
ec
D
Monthly IC
03
02
01
00
D
ec
ec
D
ec
99
98
97
ec
96
ec
D
ec
D
ec
ec
95
94
-40%
12M Average IC
IC Horizon
Average IC
3.6%
0.7%
-0.3%
-1.9%
-1.8%
1 month
3 month
6 month
12 month
24 month
4.0%
2.0%
0.0%
-2.0%
-4.0%
1
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
IC Decay
Accumulated IC
Long Short Q1-Q5

Last 5
Last 3
Last Year
Years
Years
1.4%
3.7%
2.0%
4.7%
-4.7%
-6.0%
-6.3%
-8.6%
5.9%
9.7%
7.9%
12.3%
10.2%
10.9%
13.1%
20.1%
0.58
0.88
0.60
0.61
1940%
1957%
1962%
1965%
10
All History
Top Quintile
LS Return
08
07
D
ec
D
ec
06
05
D
ec
D
ec
04
03
D
ec
D
ec
08
D
ec
07
06
05
D
ec
D
ec
D
ec
04
03
02
D
ec
2500%
1940%
1957%
1962%
1965%
2000%
Turnover
All History
D
ec
D
ec
01
00
99
D
ec
98
D
ec
D
ec
D
ec
97
96
95
D
ec
D
ec
D
ec
Last 3
Last 1 Year
Years
7.9%
12.3%
13.1%
20.1%
0.60
0.61
1962%
1965%
94
5.9%
10.2%
0.58
1940%
Bottom Quintile
20%
15%
10%
5%
0%
-5%
-10%
-15%
-20%
Long Short Returns
Last 5
Years
9.7%
10.9%
0.88
1957%
All History
D
ec
02
01
D
ec
D
ec
D
ec
00
99
98
D
ec
D
ec
97
96
D
ec
D
ec
D
ec
D
ec
95
94
1500%
1000%
500%
Turnover
08
D
ec
ec
07
06
ec
D
ec
ec
05
04
03
02
ec
ec
ec
ec
D
01
00
99
98
ec
ec
D
ec
96
D
ec
95
ec
94
ec
D
97
0%
Pure Factor Return

Pure Factor Sharpe
2.84%
2.74%
1.04
Last 5
Years
3.18%
3.00%
1.06
Last 3
Last 1 Year
Years
3.68%
5.38%
3.63%
5.29%
1.02
1.02
8%
6%
Factor Return
All History
4%
2%
0%
-2%
Factor Return
08
D
ec
07
06
ec
D
05
ec
D
ec
04
03
ec
D
02
ec
D
ec
D
ec
01
00
99
ec
D
ec
98
97
ec
D
ec
D
ec
95
D
ec
94
ec
D
96
-4%
30 April 2009
26
Relative Strength Indicator - 14Days - Time Series Normalised - continued

Fractiles
Informatio
Excess
Volatility
n Ratio
Return
1.1%
16.9%
0.07
0.8%
16.1%
0.05
1.6%
16.1%
0.10
-0.2%
16.2%
-0.01
-3.5%
14.7%
-0.24
10
Cumulative Return
Q1
Q2
Q3
Q4
Q5
Market
Annual
Return
11.7%
11.4%
12.2%
10.3%
7.1%
10.6%
Q1
All History
Q1
Q2
Q3
Q4
Q5
52.4%
50.4%
49.8%
49.1%
47.3%
Last 5
Years
51.7%
51.9%
50.4%
49.5%
46.1%
Last 3
Last 1 Year
Years
51.5%
51.9%
50.9%
52.0%
50.8%
49.2%
50.0%
49.6%
46.4%
46.8%
Q2
Q3
Q5
08
07
D
ec
D
ec
06
05
D
ec
D
ec
04
03
Q4
D
ec
02
D
ec
D
ec
01
00
D
ec
D
ec
D
ec
99
98
97
D
ec
96
D
ec
D
ec
D
ec
D
ec
95
94
Market
60.0%
58.0%
56.0%
54.0%
52.0%
50.0%
48.0%
46.0%
44.0%
42.0%
40.0%
Q1
Q2
All History
Q3
Last 5 Years
Q4
Last 3 Years
Q5
Last 1 Year
Average IC
Large
Small
All
Std Dev IC
Small
Large
All
All History
5.34%
2.52%
3.63%
All History
13.88%
11.56%
10.35%
Last 5
Years
6.28%
3.07%
4.18%
Last 3
Latest No. of
Last 1 Year
Years
Stocks
94
6.22%
8.34%
190
1.78%
1.84%
284
3.22%
3.58%
Last 5
Years
15.00%
10.12%
9.20%
Last 3
Last 1 Year
Years
16.53%
19.91%
9.09%
11.97%
9.50%
12.33%
Last 5
Years
Last 3
Years
Size
9.0%
8.0%
7.0%
6.0%
5.0%
4.0%
3.0%
2.0%
1.0%
0.0%
Large
All History
Small
Last 3 Years
Last 5 Years
All
Last 1 Year
All History
Seasonality
Utilities
Telecomm.
Services
Information
Tech
Financials
Health Care
Consumer
Staples
em
be
D
ec
% of Market Covered
08
D
ec
D
ec
07
06
D
ec
04
03
05
D
ec
D
ec
02
Stocks with Factor

D
ec
01
D
ec
00
D
ec
99
D
ec
96
97
98
D
ec
D
ec
N
ov
em
be
ct
ob
er
O
Se
p
Au
te
m
be
r
gu
st
ly
Ju
Ju
n
ay
r il
Ap
ch
ar
M
ry
Fe
br
ua
r
nu
a
99%
D
ec
99%
95
99%
D
ec
99%
94
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
400
350
300
250
200
150
100
50
0
D
ec
Last 3
Last 1 Year
Years
320
317
Number of Stocks
228
Last 5
Years
290
15
15
15
14
14
14
14
14
14
14
14
14
Factor Coverage
All History
Last 3 Years
10%
9%
8%
7%
6%
5%
4%
3%
2%
1%
0%
Samples
Ja
January
February
March
April
May
June
July
August
September
October
November
December
Information
IC Std Dev
Coefficient
7.5%
11.5%
0.6%
11.3%
3.3%
9.6%
4.1%
8.9%
3.2%
10.6%
6.0%
12.2%
9.1%
11.2%
0.7%
8.4%
2.4%
9.4%
4.4%
10.5%
0.2%
9.3%
1.9%
10.4%
Last 5 Years
D
ec
5.09%
0.38%
4.87%
2.42%
4.73%
7.96%
6.29%
16.04%
5.56%
7.41%
Consumer
Discretionary
1.73%
0.88%
6.06%
2.80%
4.16%
9.52%
4.94%
7.20%
12.50%
8.01%
Industrials
-0.61%
3.21%
4.85%
5.46%
5.48%
1.24%
7.10%
0.42%
1.44%
4.51%
18%
16%
14%
12%
10%
8%
6%
4%
2%
0%
-2%
Materials
Energy
Materials
Industrials
Consumer Staples
Health Care
Financials
Information Tech
Telecomm. Services
Utilities
Latest
Number of
Stocks
32
56
45
31
16
10
73
3
3
15
Energy
All History
Sector
30 April 2009
27
Important disclosures:
Recommendation definitions
Volatility index definition*
Financial definitions
Macquarie - Australia/New Zealand

Outperform return >5% in excess of benchmark return
Neutral return within 5% of benchmark return
Underperform return >5% below benchmark return
This is calculated from the volatility of historical

price movements.
All "Adjusted" data items have had the following

adjustments made:
Added back: goodwill amortisation, provision for
catastrophe reserves, IFRS derivatives & hedging,
IFRS impairments & IFRS interest expense
Excluded: non recurring items, asset revals, property
revals, appraisal value uplift, preference dividends &
minority interests
Macquarie Asia/Europe
Outperform expected return >+10%
Neutral expected return from -10% to +10%
Underperform expected return <-10%
Very highhighest risk Stock should be

expected to move up or down 60100% in a year
investors should be aware this stock is highly
speculative.
High stock should be expected to move up or
down at least 4060% in a year investors should
be aware this stock could be speculative.
Macquarie First South - South Africa

Outperform expected return >+10%
Neutral expected return from -10% to +10%
Underperform expected return <-10%
Macquarie - Canada
Outperform return >5% in excess of benchmark return
Neutral return within 5% of benchmark return
Underperform return >5% below benchmark return
Macquarie - USA
Outperform (Buy) return >5% in excess of benchmark
return (Russell 3000)
Neutral (Hold) return within 5% of benchmark return
(Russell 3000)
Underperform (Sell) return >5% below benchmark
return (Russell 3000)
Lowmedium stock should be expected to move

up or down at least 2530% in a year.
EPS = adjusted net profit / efpowa*

ROA = adjusted ebit / average total assets
ROA Banks/Insurance = adjusted net profit /average
total assets
ROE = adjusted net profit / average shareholders funds
Gross cashflow = adjusted net profit + depreciation
*equivalent fully paid ordinary weighted average
number of shares
Low stock should be expected to move up or

down at least 1525% in a year.
* Applicable to Australian/NZ/Canada stocks only
All Reported numbers for Australian/NZ listed stocks

are modelled under IFRS (International Financial
Reporting Standards).
Medium stock should be expected to move up or

down at least 3040% in a year.
Recommendations 12 months
Note: Quant recommendations may differ from
Fundamental Analyst recommendations
Recommendation proportions For quarter ending 31 March 2009

Outperform
Neutral
Underperform
AU/NZ
40.44%
38.60%
20.96%
Asia
49.55%
15.57%
34.88%
RSA
44.83%
39.66%
15.52%
USA
38.49%
46.43%
15.08%
CA
67.19%
28.12%
4.69%
EUR
43.84%
39.04%
17.12%
Analyst Certification: The views expressed in this research accurately reflect the personal views of the analyst(s) about the subject securities or
issuers and no part of the compensation of the analyst(s) was, is, or will be directly or indirectly related to the inclusion of specific recommendations or
views in this research. The analyst principally responsible for the preparation of this research receives compensation based on overall revenues of
Macquarie Group Ltd ABN 94 122 169 279 (AFSL No. 318062 )(MGL) and its related entities (the Macquarie Group) and has taken reasonable care to
achieve and maintain independence and objectivity in making any recommendations.
Disclaimers: Macquarie Securities (Australia) Ltd; Macquarie Capital (Europe) Ltd; Macquarie Capital Markets Canada Ltd; Macquarie Capital Markets
North America Ltd; Macquarie Capital (USA) Inc; Macquarie Capital Securities Ltd; Macquarie Capital Securities (Singapore) Pte Ltd; Macquarie
Securities (NZ) Ltd; and Macquarie First South Securities (Pty) Limited are not authorised deposit-taking institutions for the purposes of the Banking Act
1959 (Commonwealth of Australia), and their obligations do not represent deposits or other liabilities of Macquarie Bank Limited ABN 46 008 583 542
(MBL) or MGL. MBL does not guarantee or otherwise provide assurance in respect of the obligations of any of the above mentioned entities. MGL
provides a guarantee to the Monetary Authority of Singapore in respect of the obligations and liabilities of Macquarie Capital Securities (Singapore) Pte
Ltd for up to SGD 35 million. This research has been prepared for the general use of the wholesale clients of the Macquarie Group and must not be
copied, either in whole or in part, or distributed to any other person. If you are not the intended recipient you must not use or disclose the information in
this research in any way. Nothing in this research shall be construed as a solicitation to buy or sell any security or product, or to engage in or refrain
from engaging in any transaction. In preparing this research, we did not take into account the investment objectives, financial situation and particular
needs of the reader. Before making an investment decision on the basis of this research, the reader needs to consider, with or without the assistance of
an adviser, whether the advice is appropriate in light of their particular investment needs, objectives and financial circumstances. There are risks
involved in securities trading. The price of securities can and does fluctuate, and an individual security may even become valueless. International
investors are reminded of the additional risks inherent in international investments, such as currency fluctuations and international stock market or
economic conditions, which may adversely affect the value of the investment. This research is based on information obtained from sources believed to
be reliable but we do not make any representation or warranty that it is accurate, complete or up to date. We accept no obligation to correct or update
the information or opinions in it. Opinions expressed are subject to change without notice. No member of the Macquarie Group accepts any liability
whatsoever for any direct, indirect, consequential or other loss arising from any use of this research and/or further communication in relation to this
research.
Other Disclaimers: In Canada, securities research is prepared, approved and distributed by Macquarie Capital Markets Canada Ltd, a participating
organisation of the Toronto Stock Exchange, TSX Venture Exchange & Montral Exchange. Macquarie Capital Markets North America Ltd., which is a
registered broker-dealer and member of FINRA, accepts responsibility for the contents of reports issued by Macquarie Capital Markets Canada Ltd in
the United States and to US persons and any person wishing to effect transactions in the securities described in the reports issued by Macquarie
Capital Markets Canada Ltd should do so with Macquarie Capital Markets North America Ltd. Securities research is issued and distributed by
Macquarie Securities (Australia) Ltd (AFSL No. 238947) in Australia, a participating organisation of the Australian Securities Exchange; Macquarie
Securities (NZ) Ltd in New Zealand, a licensed sharebroker and New Zealand Exchange Firm; Macquarie Capital (Europe) Ltd in the United Kingdom,
which is authorised and regulated by the Financial Services Authority (No. 193905); Macquarie Capital Securities Ltd in Hong Kong, which is licensed
and regulated by the Securities and Futures Commission; Macquarie Capital Securities (Japan) Limited in Japan, a member of the Tokyo Stock
Exchange, Inc., Osaka Securities Exchange Co. Ltd, and Jasdaq Securities Exchange, Inc. (Financial Instruments Firm, Kanto Financial Bureau(kinsho) No. 231, a member of Japan securities Dealers Association and Financial Futures Association of Japan); Macquarie First South Securities (Pty)
Limited in South Africa, a member of the JSE Limited and in Singapore, Macquarie Capital Securities (Singapore) Pte Ltd (Company Registration
Number: 198702912C), a Capital Markets Services licence holder under the Securities and Futures Act to deal in securities and provide custodial
services in Singapore. Pursuant to the Financial Advisers (Amendment) Regulations 2005, Macquarie Capital Securities (Singapore) Pte Ltd is exempt
from complying with sections 25, 27 and 36 of the Financial Advisers Act. Clients should contact analysts at, and execute transactions through, a
Macquarie Group entity in their home jurisdiction unless governing law permits otherwise. Macquarie Capital (USA) Inc., which is a registered brokerdealer and member of FINRA, accepts responsibility for the content of each research report prepared by one of its non-US affiliates when the research
report is distributed in the United States by Macquarie Capital (USA) Inc. Macquarie Capital (USA) Inc. affiliate research reports and affiliate employees
are not subject to the disclosure requirements of FINRA rules. Any persons receiving this report directly from Macquarie Capital (USA) Inc. and wishing
to effect a transaction in any security described herein should do so with Macquarie Capital (USA) Inc. The information contained in this document is
confidential. If you are not the intended recipient, you must not disclose or use the information in this document in any way. If you received it in error,
please tell us immediately by return e-mail and delete the document. We do not guarantee the integrity of any e-mails or attached files and are not
responsible for any changes made to them by any other person. MGL has established and implemented a conflicts policy at group level (which may be
revised and updated from time to time) (the "Conflicts Policy") pursuant to regulatory requirements (including the FSA Rules) which sets out how we
must seek to identify and manage all material conflicts of interest. Disclosures with respect to the issuers, if any, mentioned in this research are
available at www.macquarie.com/research/disclosures. Macquarie Group
Auckland
Tel: (649) 377 6433
Bangkok
Tel: (662) 694 7999
Calgary
Tel: (1 403) 218 6650
Hong Kong
Tel: (852) 2823 3588
Jakarta
Tel: (62 21) 515 1818
Johannesburg
Tel: (2711) 583 2000
Kuala Lumpur
Tel: (60 3) 2059 8833
London
Manila
Tel: (44 20) 3037 4400 Tel: (63 2) 857 0888
Melbourne
Tel: (613) 9635 8139
Montreal
Tel: (1 514) 925 2850
Mumbai
Perth
Tel: (91 22) 6653 3000 Tel: (618) 9224 0888
Seoul
Tel: (82 2) 3705 8500
Shanghai
Singapore
Tel: (86 21) 6841 3355 Tel: (65) 6231 1111
Sydney
Tel: (612) 8232 9555
Taipei
Tokyo
Tel: (886 2) 2734 7500 Tel: (81 3) 3512 7900
Toronto
Tel: (1 416) 848 3500
New York
Tel: (1 212) 231 2500
Available to clients on the world wide web at www.macquarie.com/research and through Thomson Financial, FactSet, Reuters and Bloomberg.
30 April 2009
28
Notes
30 April 2009
29
Research
Heads of Equity Research
David Rickards (Global Co Head)
John OConnell (Global Co Head)
Industrials
(852) 2823 3538
(612) 8232 7544
Consumer Staples
Food & Beverages
Greg Dring
(612) 8232 3104
(612) 8232 3172
Retailing
Warren Doak (New Zealand)
Greg Dring
(649) 363 1416

(612) 8232 3104
(612) 8232 8531
Financials
Banks
Ben Zucker
Tom Quarmby
(612) 8232 6089

(612) 8232 8668
Diversified Financials
Deana Mitchell
(612) 8232 4576
Insurance
Tony Jackson
Deana Mitchell
(612) 8232 4442

(612) 8232 4576
(612) 8232 4130
Industrials

Ian Myles
(649) 363 1416

(612) 8232 4157
(649) 363 1416

(612) 8232 3104
(612) 8232 8633
Adam Simpson
Benn Skender
Andrew Wackett
(612) 8232 4439

(612) 8232 6846
(618) 9224 0867
Quantitative
Martin Emery (Hong Kong)
Scott Hamilton
George Platt
(852) 2823 3582

(612) 8232 3544
(612) 8232 6539
Data Services (Australia & New Zealand)
Chemicals/Containers, Packaging/Paper &

Forest Products, Construction Materials
Stephen Hudson (New Zealand)
John Purtell
(649) 363 1414

(612) 8232 8633
Jim Copland
Len Eldridge
Brendan Harris
Sophie Spartalis
(612) 8232 0397

(618) 9224 0838
(612) 8232 3575
(612) 8232 5159
Sheridan Duffy
(612) 8232 9786
Economics and Strategy

Tanya Branwhite (Strategy)
Richard Gibbs (Head of Economics)
Neale Goldston-Morris (Strategy)
Brian Redican (Aus Economics)
Mark Tierney (Intl Economics)
(612) 8232 7628

(612) 8232 3935
(612) 8232 7562
(612) 8232 7016
(612) 8232 3121
Find our research at
Real Estate
Property Trusts & Developers
Callum Bramah
Paul Checchin
(612) 8232 7647

(612) 8232 4197
Telecommunications
Andrew Levy
(612) 8232 5165
Macquarie:
www.macquarie.com.au/research
Thomson:
www.thomson.com/financial
Reuters:
www.knowledge.reuters.com
Bloomberg:
MAC GO
Factset:
http://www.factset.com/home.aspx
Contact Gareth Warfield for access (612) 8232 3207
See and hear our analysts at
Stephen Hudson (New Zealand)

Gavin Maher
(649) 363 1414

(612) 8232 4151
Commodities & Precious Metals
Capital Goods
Greg Dring
John Purtell
Transportation - Infrastructure
Utilities
Healthcare
Steve Wheen
(649) 363 1416

(612) 8232 4157
(612) 8232 7124
Global Metals & Mining
Energy
Adrian Wood

Ian Myles
Russell Shaw
Materials
Media and Tourism & Leisure

Alex Pollak
Emerging Leaders
Transportation /Commercial Services
Jim Lennon (London)

Adam Rowley (London)
(44 20) 3037 4271

(44 20) 3037 4272
www.macquarie.com.au/macquariedigital
Toll free from overseas

Canada
Hong Kong
Japan
New York
Singapore
1800 989 8159

800 96 2049
0053 161 6437
1888 622 7862
800 616 1037
Email addresses
FirstName.Surname@macquarie.com
eg. David.Rickards@macquarie.com
Sales
Equities
Martin Dacron (Australia)
Scott Dolling (Asia)
Rob Fabbro (Continental Europe)
Charles Nelson (UK)
Duane ODonnell (Melbourne)
Dave Roberton (New Zealand)
Luke Sullivan (New York)
Stevan Vrcelj (Head of Global Sales)
Specialist Sales
(612) 8232 7421
(852) 2823 3705
(44 20) 7065 2031
(44 20) 7065 2032
(613) 9635 9183
(649) 363 1498
(1 212) 231 2507
(612) 8232 5999
Treasury & Commodities
Matthew Clegg (Index)

Andrew Mouat (Property Trusts)
George Platt (Quantitative)
Phil Zammit (Emerging Leaders)
(612) 8232 5653

(612) 8232 3151
(612) 8232 6539
(612) 8232 3122
Alternative Strategies
Anthony Panaretto (Sales)
Shannon Donohoe
(Stock borrow & loan)
Mark Donnelly (Equity finance)
Cameron Duncan (Converts)
Anthony Hourigan (Derivatives)
April 09
(612) 8232 4500

(612) 8232 6997
(612) 8232 7664
(612) 8232 7405
(612) 8232 9884
Gavin Bradley (Metals & Mining)

Matthew Forgham
(Metals & Mining)
Emma Winspear (Futures)
James Mactier (Metals & Mining)
Ian Miller (Futures)
Will Richardson (Foreign Exch)
(612) 8232 4248

(44 20) 3037 4615
(613) 9635 8275
(618) 9224 0612
(612) 8232 3555
(612) 8232 4777
Syndication
Peter Curry
Paul Staines
(612) 8232 4039

(612) 8232 7781

Macquarie Technicals PDF

Diunggah oleh

Informasi Dokumen

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Macquarie Technicals PDF

Diunggah oleh

Hak Cipta:

Format Tersedia

Quantitative analysis

Lets get technical

Technical indicators add value in Australia

Volatile markets present opportunity to use technicals

Williams %R the most powerful signal

Size matters technicals work best in large cap stocks

Market instability hurts fundamental

Using Technical Indicators as Quant

mean reverting technical indicators over trend following strategies.

Performance of Technical Indicators in

Appendices Factor definitions

mean trend Technical indicators and favours momentum based strategies.

Improving the investment process with technicals

as a knock out signal in a standard quant model.

Macquarie Research Equities - Report

Lets get technical

Defining technical indicators

Williams %R the standout technical indicator

Fig 1 Williams %R information coefficient

Fig 2 Technical performance in large vs. small caps

Small Cap - ex ASX100

Source: Macquarie Research, April 2009

Other key technical indicators that also performed well were:

Commodity Channel Index

Relative Strength Indicator

Large caps mean revert; small caps trend follow

Adding value to a quant process

Macquarie Research Equities - Report

Market instability hurts fundamental factors

Fig 3 ASX200 cross sectional volatility

Performance of value factors

Source:Macquarie Research, April 2009

Source: Macquarie Research, April 2009

Turning our attention to technicals

Davies, C. Quantamentals: Do technicals add value?, Macquarie Research Equities, (2009)

Macquarie Research Equities - Report

Using technical indicators as Quant factors

Technical indicators tested

Price Based Technical Indicators

Price & volume Based Technical Indicators

Volume Based Technical Indicators

Relative Strength Index

Money Flow Index

Source: Macquarie Research, April 2009

We test the standard

Macquarie Research Equities - Report

Performance of technical indicators in

Q1-Q5 Long Short

Source: Macquarie Research, April 2009

Macquarie Research Equities - Report

Information coefficient of mean reverting technical indicators

Source:Macquarie Research, April 2009

Macquarie Research Equities - Report

Fig 8 Williams %R information coefficient

Fig 9 Q1-Q5 long / short performance

Source:Macquarie Research, April 2009

12M Rolling LS returns

Source: Macquarie Research, April 2009

Davies, C. Quantamentals: Do Technicals add value?, Macquarie Research Equities, (2009)

Macquarie Research Equities - Report

Fig 10 Technical indicator information coefficient in

Fig 11 Q1-Q3 L/S information ratio before costs in

Small Cap - ex ASX100