Internship Programme
2010 – Statistical Analysis
and Identification of Lead
Indicators of core sectors
of the Indian Economy
Steel Industry
Submitted by : Aayush Kamani,
NMIMS,
Roll no. 101.
1
Objective
Issues of growth in the iron and steel sector as well as in other energy intensive industries
in India have been discussed from various perspectives. Historical estimates vary from
indicating an improvement to a decline in the sector’s growth. The variation depends
mainly on the time period considered, the source of data, the type of indices and
econometric specifications used for reporting productivity and growth.
India's iron and steel industry contributes about 2% of gross domestic product, or about
USD 20 billion to the country's USD 1 trillion economy. This report will analyze and
identify the lead indicators that are useful in forecasting the future outlook of the steel
industry. Additionally, this report will statistically prove a relation between these
forecasting factors i.e. lead indicators of the steel industry and the output of the steel
industry in India over the years.
2
Index
Objective……………………………………………………… 2
Methodology……………………………………..…………… 4
Forecasting factors…………………………………………... 7
Results………………………………………………………... 9
Interpretation of results……………………………………... 12
Conclusion………………………………………..………… 14
Acknowledgements………………………………………… 15
Bibliography……………………………………………….. 16
10. Annexure……………………………………………………. 17
3
Methodology
The study was conducted and presented to understand and highlight the main quantitative
lead parameters that can be used to forecast and predict the state of the industry in the
future.
A lead indicator is a measurable parameter or metric, which will help to predict or give
prior information about the occurrence of some related critical events.
4
The Indian Steel Industry
India has now emerged as the fifth largest producer of steel in the world with a
production capacity of 63MT, while on the other hand some Asian countries like Japan,
south Korea saw significant decline in steel production. Almost all varieties of steel is
now produced in India.
The growth of the steel industry in India is dependant, to a large extent, on the level of
consumption of steel in the domestic market. Steel consumption is significant in housing,
auto manufacturing and infrastructure.
In recent years the surge in housing industry of India has led to increase in the domestic
demand for steel. It is estimated that by 2011 the demand for houses would grow to
around 400 million units. This will necessitate a minimum outlay of US$ 890 billion. In
India, demand is being driven by infrastructure projects, like construction of roads,
railways and ports. With salaries rising, consumers are becoming more discerning with
regard to their cars, houses, which will boost steel demand.
The private sector contribution in the total output has been increasing in India. During the
last decade more than 12mt of capacity has been added in the steel industry, this is mostly
in the private sector. With increasing need for large investments in the industry private
sector’s role would be crucial in the development of the steel industry. 5-6 major players
such as SAIL, TATA, JSW and ESSAR to name a few dominate the industry. Significant
capacity expansion to the tune of 50MT is expected in the next decade.
Growth in Future:
• Tata Steel ranks 5th in the world steel production and the company have plans of
expanding its capacity from 6.8mt to 10mt by the year 2011
• SAIL, India's biggest producer of steel has plans of increasing the production to
24.98 million tons annually
• The acquisition of the Corus, the Anglo-Dutch steel manufacturer by the Tata
Steel
• The Algoma Steel, Canada was acquired by Essar Global for US$ 1.63 billion
• JSW steel has plans of expanding capacity from 7.8mt to 11mt by 2011
5
India has a potential for exponential growth in steel consumption…
6
Forecasting factors
Gross Domestic Product (GDP) - is a measure of a country's overall economic output. It
is the market value of all final goods and services made within the borders of a country in
a year. GDP is widely used by economists to gauge the health of an economy, as its
variations are relatively quickly identified. However, its value as an indicator for the
standard of living is considered to be limited. GDP is not a tool of economic projections,
which would make it subjective; it is just a measurement of economic activity. That is
why it does not measure what is considered the sustainability of growth. A country may
achieve a temporarily high GDP by over-exploiting natural resources or by misallocating
investment.
GDP.xls
Gross Fixed Capital Formation (GFCF) - Gross fixed capital formation (GFCF) is a
macroeconomic concept used in official national accounts. Statistically it measures the
value of acquisitions of new or existing fixed assets by the business sector, governments
and "pure" households (excluding their unincorporated enterprises) less disposals of fixed
assets. GFCF is a component of the expenditure on GDP and thus shows something about
how much of the new value added in the economy is invested rather than consumed.
Fluctuations in this indicator are often considered to show something about future
business activity, business confidence and the pattern of economic growth. In times of
economic uncertainty or recession, typically business investment in fixed assets will be
reduced, since it ties up additional capital for a longer interval of time with a risk that it
will not pay off. Conversely, in times of robust economic growth, fixed investment will
increase across the board, because the observed market expansion makes it likely that
such investment will be profitable in the future.
GFCF.xls
(refer Annexure I)
7
Auto Sales - Auto and truck Sales measure the monthly sales of all domestically
produced vehicles. They are considered an important indicator of consumer demand,
accounting for roughly 25% of total retail sales. Demand for big ticket items such as
autos and trucks tend to be interest rate sensitive, making the motor vehicle sector a
leading indicator of business cycles.
The automobile industry in India happens to be the ninth largest in the world. Following
Japan, South Korea and Thailand, in 2009, India emerged as the fourth largest exporter of
automobiles. Several Indian automobile manufacturers have spread their operations
globally as well, asking for more investments in the Indian automobile sector by the
MNCs.
autosales.xls
(refer Annexure I)
8
Results
Steel Production – Auto sales and Gross Fixed Capital Formation (GFCF)
Regression output:
Regression equation:
9
Steel Production ('000 Metric Tons)
70,000
60,000
50,000
Actual Steel
40,000
Testing for “Normality” in data sets: Production
30,000 Predicted Steel
Production
20,000
10,000
0
1 2 3 4 5 6 7 8 9 10 11
As can be seen from the regression output, we have used only 10 data sets. This is due to
limitations in finding and grouping common year data to get the appropriate output. To
test whether the existing data set tends towards a “normal distribution” we conducted the
following test.
Establishing the underlying distribution of a data set (or random variable) is crucial for
the correct implementation of some statistical procedures. For example, both the small
sample t test and ANOVA, require that the distribution of the underlying data be normal.
Therefore, we first need to establish whether the normal applies before we can correctly
implement these statistical procedures.
We used the Anderson-Darling Normality test to prove that the data used does follow a
normal distribution. For the normality test, the hypotheses are,
The Anderson-Darling methodology test's the p-value. If the p-value that is generated
from this test is greater than the α value of 0.05 then we will accept the null hypothesis
(H0) and safely say that the data follows a normal distribution.
10
N 10
AD 0.570
P-Value 0.103
Mean 71452
StDev 25700
N 10
AD 0.412
P-Value 0.273
As can be seen from the above graphs, the p-value of GFCF obtained from conducting
the Anderson-Darling test is 0.103 which greater than the α value of 0.05. Similarly, the
p-value of auto sales is 0.273 which is greater than the α value of 0.05. Hence, we will
accept the null hypothesis of this test and can safely say that this data set follows normal
distribution.
Interpretation of results
11
A. The regression output has three components:
The above gives the overall “goodness of fit” measures. The goodness of fit of a
statistical model describes how well it fits a set of observations. Measures of goodness of
fit typically summarize the discrepancy between observed values and the values expected
under the model in question.
R2 is the most important measure that signifies how much of the variation in the steel
production is explained by the independent variables i.e. both, auto sales and GFCF
combined explain 97% of the variation of steel production.
2. ANOVA Table:
The ANOVA table gives the answer for testing the claim that there is no significant
relationship between the steel production (dependent variable) and GFCF and auto sales
(independent variables). At 95% confidence level if the “Significance F” value is less
than α value (0.05) i.e. 100% - 95% = 5%, then we can reject the claim that there is no
significant relationship between the dependent and independent variables.
Significance F value
Multiple Regression 0
The P-value for individual independent variables is given. This too tests the hypothesis
that there is no relationship between the independent variable and dependent variable.
Thus, the claim is rejected that there is no relationship between the independent and
dependent variable if the P-value < α value.
12
GFCF – Steel Production 0.016
Auto Sales – Steel Production 0.040
When the focus is on the relationship between a dependent variable and one or more
independent variables. More specifically, regression analysis helps us understand how the
typical value of the dependent variable changes when any one of the independent
variables is varied, while the other independent variables are held fixed.
From the above equation and graph we can see the relationship between the dependent
variable (steel production) and independent variables (GFCF and auto sales). The value
of “y” in the equation gives us the value of steel production in thousand metric tons.
“0.016511(x1)” tells us how much of an impact GFCF has on determining the quantity of
steel production. Similarly, “0.106159(x2)” tells us how much of an impact auto sales has
on determining the quantity of steel production.
The graph is a pictorial relationship depicting the accuracy of the regression equation.
The predicted values have been generated using the equation and have been compared to
the actual values of steel production in India. As can be seen, the graph depicts a close
approximation to the real values and is a good indicator.
steel_output.xls
(refer Annexure I)
Conclusion
13
India is the fifth largest producer of steel in the world. India steel industry has grown on
account of strong domestic demand driven by investment in infrastructure and growth in
the auto sector. The scope of steel industry is huge and industry estimates indicate that
the steel industry will continue to grow substantially in the coming years.
By analyzing the above-mentioned factors we can conclude that auto sales and GFCF
play an important role in determining the future outlook of the Indian Steel Industry.
GFCF is a more focused approach than GDP as it covers investments made for
infrastructure development. The growth in auto sales helps in predicting the future
demand of steel as a major component for automobile manufacturing is steel.
Some of the important indicators to emphasize the robustness of the steel industry:
Acknowledgements
14
I thank ICICI Bank for providing me with this opportunity to understand and analyze the
Steel Industry in India. Steel being one of the basic industries is of significant importance
for the overall performance of the country and economy. By conducting this study I have
gained a lot of understanding into the functioning and parameters that affect the steel
industry in India.
I express my heartfelt gratitude to Mr. Vivek Sharma of the Global Risk Management
Group who has been kind enough to guide me through the whole process and without
whose adept suggestions this project would not have been possible.
I also take this opportunity to thank Mr. Deepak Kenkre of the Global Risk Management
Group whose insight into the industry was most incisive and whose timely feedback
shaped the project the way it is.
Lastly, I would like to thank my family and friends who have extended their utmost
support during the course of this project.
Bibliography
15
1. Ministry of Finance Website – www.finmin.nic.in
2. Filliben, J. J. (February 1975). "The Probability Plot Correlation Coefficient Test
for Normality"
3. Bloomberg
4. The Associated Chambers of Commerce and Industry of India (ASSOCHAM)
5. India Statistics – www.indiastat.com
6. CRISIL Report
7. ICICI Securities Report
8. World Steel
16
Annexure
Annexure I
17