Table
of
Contents
1.0 INTRODUCTION ....................................................................................... 1 1.1 BACKGROUND OF THE PLUS COMPANY ........................................ 1 1.2 OVERVIEW OF MALAYSIA ECONOMY.............................................. 1 2.0 QUANTITATIVE TECHNIQUE ANALYSIS ................................................ 2
2.1 RELATIONSHIP BETWEEN COMPANYS SALES AND GDP ..................... 2 2.2 GDP FORECASTING ............................................................................. 4 2.3 SALES FORECASTING .......................................................................... 6
3.0 INTERPRETATION ABOUT THE FORECASTING ................................... 6 4.0 CONCLUSION ........................................................................................... 8 5.0 REFERENCES .......................................................................................... 9 6.0 APPENDIX ............................................................................................... 10
1.0 INTRODUCTION
PLUS was incorporated in Malaysia on 27 June 1986 as Highway Concessionairs Berhad and adopted its present name on 13 May 1988. On 25 May 1988, 25.0 million ordinary shares of RM 1.00 each in PLUS were issued to UEM and the subscriber shares were transferred to UEM, whereupon PLUS became a wholly-owned subsidiary of UEM. PLUS core business consists of the design, construction, financing, operation and maintenance of certain toll roads and expressways in Malaysia.
Malaysia is currently in its fourth year of economic growth following a deep recession caused by the Asian financial crisis of 1997-98. Following a 7.5% decline in real gross domestic product (GDP) in 1998, Malaysia experienced real GDP growth of 6.1% in 1999 and 8.3% in 2000. Growth slowed to 0.4% in 2001 on weakened demand for the country's manufactured exports in the face of a global economic slowdown, but it is projected to recover to 3.4% in 2002. Imports of capital goods have been rising, which has lowered the country's merchandise trade surplus.
YEAR
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
126.4
138.9
151.7
166.6
183.3
196.7
182.2
193.3
209.4
210.2
Sales (Milliom)
890
938
999
1011
1027
1076
946
1105
1153
1220
In this case, Linear Regression method will be used. So let sales as the dependent variable --y; GDP as the independent variable --x. That is to say the PLUS companys sales are affected by the GDPs changes. The relationship between the two variables is given as following equation: y = a + bx and ybx a= n b= nx2(x)2 nxyxy
According to the figures of sales and GDP showing above, the regression line should be represented as1: y = 492 +3.1x
(See Appendix1)
Scatter diagram2 The linear regression line shows in the scatter diagram as below:
y = 3.0985x + 491.56
50.0
100.0
150.0
200.0
Coefficient of correlation (r) The coefficient of gives an indication of the strength of the linear relationship between two variables. These are several possible formulae but a practice one is: nxyxy r=
(nx2(x)2)(ny2(y)2)
However wide the scatter of the data, a line of best fit can be calculated using squares. Although such a line can always be calculated, it does not follow that
1
All the relevant calculations in this assignment are attached in Appendix. This item is In order to present the diagram expressly, the scale on y axis crosses at 800 million point
attached in Appendix 1.
2
purposely.
the best fit line is likely to be much use for predictive purposes, unless it is an accurate presentation of any trend in the data. To find out how good the line of best fit really is, a measure called the coefficient of determination is calculated. This measure denoted by calculates what proportion of the variation in the actual values of may be predicted by changes in the values of x.
According to the figures and the formula above: r = 0.88 r2 =0.77 or 77%
(See Appendix 1)
Figure 0.77 shows a strong relationship between sales and GDP. This result may be interpreted that in the problem 77% of the variation in actual faulty of sales parts delivered may be predicted by change in the actual value of GDP spent on inspection.
As the name suggests, time series analysis uses some form of mathematical of statistical analysis on past data arranged in a time series. A time series typically has four components: trend (T), seasonality fluctuations (S), cycle fluctuations (C), and random variation (R). There are two models of time series model statistics: multiplicative model and additive model. The 15 years GDP figures of Malaysia show in table 3.
YEAR 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 GDP 81.0 89.2 97.2 106.0 116.1 126.4 138.9 151.7 166.6 186.3 196.7 182.2 193.3 209.4 210.2
These figures show the fluctuation of each year is constant, so the additive method should be used herein. In this case, GDP forecasting is for long term, so the seasonality fluctuation should not be considered. Therefore, the time series forecasting equation is given by: Forecast = T + C + R Trend Based on the GDP figures showing above, generally three points moving average might be used. According to the calculations, the three point moving average shows that the trend of GDP increases from RM 204.3 Billion to RM 89.1 Billion. Thus: Average Increase of GDP = 8.86 Billion RM per year Trend of GDP for 2001 = 213.16 Billion RM Trend of GDP for 2002 = 222.02 Billion RM
(See Appendix 2)
Cyclical Fluctuation As we adopted the tree points moving average, the forecasting year 2002 belongs to the period 1. The cyclical variation of 2002 is: Cyclical Variation =0.255 Billion RM
(See Appendix 2)
2002 is:
GDP = RM 221.765 Billion
(See Appendix 3)
GDP(billion)
220 190 160 130 100 70
87
88
90
90
91
92
93
94
95
96
97
98
99
00 20
19
19
19
19
19
19
19
19
19
19
19
19
19
REAL GDP
TREND GDP
Based on the above demonstrations, the sales of PLUS Company for the next year can be calculated by this formula: y= 492 + 3.1x
20
YEAR
01
only shows a strong association between the two variables. A low correlation coefficient, somewhere near zero, does not always mean that there is no relationshio between the variables. All it says that there is no linear relationship between the variablesthere may be strong relationship but of a non-linear kind.
Quantitative techniques are techniques or varying levels of statistical complexity which are based on analyzing past data of the item to be forecast eg. sales figures. However sophisticated the technique used, there is the underlying assumption that past patterns will provide some guidance to the future. Clearly for many operational items (material usage, sales of existing products) the past does serve as a guide to the future, but there are circumstances for which no data are available rg the launching of a completely new product, where other, more qualitative techniques are required. These techniques are dealt with briefly first and then the detailed quantitative material follows.
Qualitative techniques are techniques which are used when data are scarce, eg. the first introduction of a new product. The techniques use human judgment and experience to turn qualitative information into quantitative estimates. Although qualitative techniques are used for both short and long term purposes, their use becomes of increasing importance as the time scale
of the forecast lengthens. Even when past data are available, so that standard quantitative techniques can be used, longer term forecasts require judgment, intuition, experience, etc., that is, qualitative factors, to make them more useful. As the time scale lengthens, past patterns become less and less meaningful. The qualitative methods briefly dealt with in this manual are the DELPHI METHOD, MARKET RESEARCH, and HISTORICAL ANALOGY.
4.0 CONCLUSION
In short, linear regression module and time series analysis module are basically used for simple forecasting. Both of the two models use a certain historical data to establish the trend and they are the quantitative techniques, so the forecasting might be less accurate. In practice, many factors should be taken into account and many other techniques should be used. When past quantitative data are unavailable for the item to be forecast, then inevitably much more judgment is involved in making forecasts. Some of the qualitative techniques mentioned above use advanced statistical techniques.
Nevertheless, any such method may prove to be a relatively poor forecaster, purely due to the lack of appropriate quantitative data relating to the factor being forecast.
5.0 REFERENCES
1. Brown, R. G. Statistical Forecasting for Inventory Control, New York: McGraw-Hill. 2. Ashton, A.H., and Sapienza, A. Time-series Analysis: Forecasting and Control. San Francisco: Holden Day.
3. Lee, D. R. A Forecast of Lodging Supply and Demand. The Control HRA
Quarterly: pp27-40. 4. Chambers, J. C., Mullick, S. K., and Smith, D. D. How to Choose the Right Forecasting Technique. Harvard Business Review 49, 4: pp45-47. 5. Gould. D,F.J. Eppen, Gary D. Schmidt. Charles. Time-series Forecasting Models. Quantitative Concepts for Management, p151. 6. Mahmoud, E. Accuracy in Forecasting: A Summary. Journal of Forecasting.
6.0 APPENDIX
Appendix 1.
GDP Year
1,992 1,993 1,994 1,995 1,996 1,997 1,998 1,999 2,000 2,001
(bn RM)
Sales
(mn RM)
XY
112,496 130,288 151,548 168,433 188,249 211,649 172,361 213,597 241,438 256,444 1,846,503
x2
15,977 19,293 23,013 27,756 33,599 38,691 33,197 37,365 43,848 44,184 316,923
Y2
792,100 879,844 998,001 1,022,121 1,054,729 1,157,776 894,916 1,221,025 1,329,409 1,488,400 10,838,321
X
126.4 138.9 151.7 166.6 183.3 196.7 182.2 193.3 209.4 210.2 1,758.7
Y
890 938 999 1,011 1,027 1,076 946 1,105 1,153 1,220 10,365 Table 1.
According to Table 1:
nxy - xy b= nx - (x)
2 2
y- bx a= n =
y = 492 +3.1x
nxy - xy r=
so,
10
Appendix 2.
Year
GDP ( G)
Cyclical+Residual Variation ( G - T )
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
81.0 89.2 97.2 106.0 116.1 126.4 138.9 151.7 166.6 183.3 196.7 182.2 193.3 209.4 210.2
267.4 292.4 319.3 348.5 381.4 417.0 457.2 501.6 546.6 562.2 572.2 584.9 612.9
89.1 97.5 106.4 116.2 127.1 139.0 152.4 167.2 182.2 187.4 190.7 195.0 204.3
0.1 -0.3 -0.4 -0.1 -0.7 -0.1 -0.7 -0.6 1.1 9.3 -8.5 -1.7 5.1
According to Table 2, Average Cyclical Variation of each period can be calculated. The calculation process is:
Total of period 1 = (-0.4)+(-0.1)+1.1+(-1.7)= -1.1 Total of period 2 = 0.1+(-0.1)+(-0.7)+9.3+5.1= 13.7 Total of period 3 = (-0.3)+(-0.7)+(-0.6)+(-8.5)= -10.1
Average cyclical variation of period 1= (-1.1)4=-0.275 Average cyclical variation of period 2= 13.75 = 2.74 Average cyclical variation of period 3= (-10.1)4=-2.525
11
Period 1
Period 2
Period 3
-0.4
Cyclical + Resodual Variation (G-T)
Total
Average Cyclical Variation
-1.1 -0.275
Table 3.
The total of the average cyclical variation is: (-0.275) + (2.74) + (-2.525) = -0.06
We should adjust the figure equal to zero. The adjustment is as the following.
Period 1 Period 2 Period 3 -0.275 - (-0.02) = - 0.255 2.74 -(-0.02) = 2.76
-.525-(-0.02) = -2.505
Then the revised cyclical variations do total zero: -0.255 + 2.76 -2.505= 0 According to Table 2, 2002 is belongs to period 1. Thus: Cycle of 2002 GDP = - 0.255 Billion RM So, GDP of 2002 is: GDP = T + C = 222.02 + (0.255) = 221.765 Billion RM
12
Appendix 3.
As the linear regression line of GDP on Sales is: y = 492 + 3.1x The sales of PLUS in 2002 is: Sales = 492 +3.1221.765= 1179.472 Million RM
13