Project Report MacroEconomics

7/31/2019 Sensitivity of BSE
Sensex to macro-
economic
indicators
Kriti Arneja, IPMX12019
[Type here]
Pre-lude
Objective:
The objective of this research study is to evaluate (independently) significance of regression at
different interval levels between different macro-economic variables and equity markets of India,
using different statistical tools.
Data Selection:
• The proxy used for equity markets is BSE Sensex.
• Different macro-economic indicators used for the study are
o NDP at Factor Prices
o CPI
o WPI
o Gold Prices
o Lending Rates
o Crude Oil Prices
o Foreign Exchange Rate
o FDI
o Balance of Payments
Duration:
The data captured is for the past 10 years, from 2008-2009 to 2017-2018.
Source:
The data is sourced from RBI Handbook of Statistics.
Statistical Tools Used:

The following statistical tools are used to evaluate study results
• Correlation and Regression

• Hypothesis Testing, t-tests
• Skewness
• Kurtosis
• Trend Lines
The research is carried out as a project undertaken for the academic module “Macro-economics”
undertaken by faculty Dr. Chandan Sharma, during IPMX12, IIM-Lucknow, Noida Campus. Use of this
project for any other inference is not recommended.
NDP at Factor Prices and BSE Sensex
India’s economy is ranked 7th largest in the world at nominal prices and 3rd largest in the world in terms of PPP. 16% of the economy is controlled by
agriculture, 30% by manufacturing sector and 54% by services sector. The CAGR from 2008-2009 to 2017-2018 is 11.5% which is appreciable.
BSE Sensex is a stock index of 31 stocks from different sectors that represent the economy. TCS has the highest weightage in the index at 12.29% followed by
Reliance Industries (11.25%), followed by HDFC Bank Limited (9.26%). The CAGR of the index growth from 2008 -2009 to 2017-2018 is 10.11%.
Year BSE NDP FC (Rs Correlation 0.97

Sensex Billion)
2017-18 32396.83 135161.84 Coefficient of determination 0.93
2016-17 27338.22 121687.92 Significance of correlation, t-test
2015-16 26322.10 109650.22 Null Hypothesis correlation between variables is 0.
2014-15 26556.53 100145.58 Alternate Hypothesis correlation is positive
2013-14 20120.12 90272.94 Degree of Freedom 8
2012-13 18202.10 80250.24 T-statistic 10.68
2011-12 17422.88 71129.47 p-value 0.0000026 Inference
2010-11 18605.18 62059.19 A 0.01 Reject
2009-10 15585.21 52466.25 A 0.05 Reject
2008-09 12365.55 45465.22 A 0.1 Reject
Mean 21491.47 86828.89 The null hypothesis is rejected at 1% significance level and can be deduced that
Standard Deviation 6305.66 29918.90 correlation between BSE Sensex and NDP FC prices is positive and strong at 1% level
Coefficient of Variation 29% 34%
Skewness 0.36 0.20
Kurtosis -0.870 -1.071
The research is carried out as a project undertaken for the academic module “Macro-economics” undertaken by faculty Dr. Chandan Sharma, during
IPMX12, IIM-Lucknow, Noida Campus. Use of this project for any other inference is not recommended.
160000.00 35000.00
140000.00 30000.00
120000.00
25000.00
100000.00
20000.00
80000.00
15000.00
60000.00
10000.00
40000.00
5000.00 Both NDP at Factor prices and BSE Sensex have a positive skewness
20000.00
and both of them have low positive values which suggests that
0.00 0.00
2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 majority of the data sets in both the variables are centered around
the mean. The coefficient of variation is low.
NDP FC (Rs Billion) BSE Sensex
This is supported by negative kurtosis in both the data sets that
suggests that both the variables have lighter tails.
160000.00
93% of the correlation is explained and the regression line is a
140000.00 y = 4.5866x - 11743
good fit since the null hypothesis is rejected at 5% levels.
120000.00
100000.00
80000.00
60000.00
40000.00
20000.00
0.00
0.00 5000.00 10000.00 15000.00 20000.00 25000.00 30000.00 35000.00
CPI and BSE Sensex
The Consumer Price Index (CPI) is a measure that examines the weighted average of prices of a basket of consumer goods and services, such as
transportation, food, and medical care. It is calculated by taking price changes for each item in the predetermined basket of goods and averaging them. The
following table shows CPI with the base year 2012. Food and beverages form 45% of the total weights, followed by Housing at 10%. Miscellaneous accounts
28% of the weights.
Year BSE Sensex CPI Correlation 0.95

2015-16 26322.10 124.70 Null Hypothesis correlation between variables is 0
2012-13 18202.10 102.70 T-statistic 6.75
2011-12 17422.88 93.30 P-value 0.000543 Inference
2010-11 - - A 0.01 Reject
2009-10 - - A 0.05 Reject
2008-09 - - A 0.1 Reject
Mean 24051.25 116.74

Standard Deviation 5560.09 15.01
Coefficient of The null hypothesis is rejected at 1% significance level and can be deduced that correlation
Variation
23% 13%
between BSE Sensex and CPI index is positive and strong at 1% level
Skewness 0.13 -0.46
Kurtosis -1.311 -0.893
160.00 35000.00
140.00 30000.00
120.00
25000.00
100.00
20000.00
80.00
15000.00
60.00
10000.00 While BSE Sensex have a positive skewness with low positive
40.00
5000.00 values which suggests that majority of the data sets in both the
20.00
variables are centered around the mean, CPI has a negative
0.00 0.00
skewness with low negative value that suggests that the data is
2012 2013 2014 2015 2016 2017 2018
skewed towards the left. The coefficient of variation is low for both
CPI BSE Sensex the variables.
Both the variables have negative kurtosis that suggests that both
the variables have lighter tails.
160.00
140.00 y = 0.0026x + 55.103 90% of the correlation is explained and the regression line is a
120.00
good fit since the null hypothesis is rejected at 5% levels.
100.00
80.00
60.00
40.00
20.00
0.00
0.00 5000.00 10000.00 15000.00 20000.00 25000.00 30000.00 35000.00
WPI and BSE Sensex
Wholesale price index comprises as far as possible all transactions at first point of bulk sale in the domestic market. Provisional monthly WPI for All
Commodities is released on 14th of every month (next working day, if 14th is holiday). The provisional index is made final after a period of eight weeks/ two
months. Manufacturing Products form 64% of the weights assigned to the goods.
Year BSE Sensex WPI Correlation 0.70

2012-13 18202.10 106.9 T-statistic 1.97
2011-12 - - p-value 0.059 Inference
2010-11 - - A 0.01 Accept
2009-10 - - A 0.05 Accept
2008-09 - - A 0.1 Reject
Mean 25155.98 111.56

Standard Deviation 5181.31 2.89 The null hypothesis is rejected at 10% significance level but is not rejected at 5% and 1% significance
Coefficient of Variation 21% 3% levels and it can be deduced that correlation between BSE Sensex and WPI is positive and but not
Skewness -0.15 -0.75 strong.
Kurtosis -0.657 -0.006
116.0 35000.00
114.0 30000.00
112.0 25000.00
110.0 20000.00
108.0 15000.00
106.0 10000.00
While BSE Sensex has skewness with low negative values which
104.0 5000.00
suggests that majority of the data set is centered around the
102.0 0.00 mean, WPI has a high negative skewness that suggests that the
2013 2014 2015 2016 2017 2018 data is skewed towards the left. The coefficient of variation is low
WPI BSE Sensex for WPI but higher for BSE Sensex
116.0
49% of the correlation is explained and the regression line is a not
115.0
y = 0.0004x + 101.71 a good fit since the null hypothesis is not rejected at 5% levels.
114.0
113.0
112.0
111.0
110.0
109.0
108.0
107.0
106.0
0.00 5000.00 10000.00 15000.00 20000.00 25000.00 30000.00 35000.00
Gold Prices and BSE Sensex
Gold Prices are determined highly by how USD as a currency performs internationally. At the local level, The Indian Bullion Jewellers Association plays a key role
in determining day to day gold rates. IBJA members include the biggest gold dealers in India, who have a collective hand in establishing prices. The IBJA then
gets into the act of determining prices by speaking to the ten biggest gold dealers. These dealers give their respective ‘buy’ and ‘sell’ quotes, depending on the
rate at which they purchased gold. IBJA then takes the average of these ‘buy’ and ‘sell’ quotes and determines the gold rate for a particular day based on this
average. This average rate is adjusted for local taxes and a rate fixed accordingly. CAGR of the gold prices for the last 10 years is 8.5%.
Year BSE Sensex Gold (Rs/10 gms) Correlation 0.70

2012-13 18202.10 30163.93 T-statistic 2.77
2010-11 18605.18 19227.08 A 0.01 Accept
2009-10 15585.21 15756.09 A 0.05 Reject
2008-09 12365.55 12889.74 A 0.1 Reject
Mean 21491.47 24586.38

Standard Deviation 6305.66 6297.36 The null hypothesis is rejected at 10% and 5% significance levels but cannot be arejected
Coefficient of Variation 29% 26% at 1% significant levels. However, it can be appreciated that correlation between BSE
Skewness 0.36 -1.06 Sensex and Gold prices is positive and strong at 1% level
Kurtosis -0.870 -0.415
35000.00 35000.00
30000.00 30000.00
25000.00 25000.00
20000.00 20000.00
15000.00 15000.00
10000.00 10000.00 While BSE Sensex has skewness with low positive values which
5000.00 5000.00
mean, Gold Prices have a high negative skewness that suggests
0.00 0.00 that the data is skewed towards the left. The coefficient of
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
variation is moderate for both BSE Sensex and Gold Prices.
Gold (Rs/10 gms) BSE Sensex
35000.00
y = 0.6984x + 9577.2
30000.00
49% of the correlation is explained and the regression line is a not
25000.00 a good fit. However, the null hypothesis is rejected at 5% levels.
20000.00
15000.00
10000.00
5000.00
0.00
0.00 5000.00 10000.00 15000.00 20000.00 25000.00 30000.00 35000.00
Gold (Rs/10 gms) Linear (Gold (Rs/10 gms))
Lending Rates and BSE Sensex
Lending rates in India are largely determined by a few RBI tools such as Repo rate, Reverse Repo Rate and the MCLR. The Repo rate, at which the RBI lends to
bank for short term purposes, as on July 31st 2019 is 5.75% while the Reverse Repo Rate, at which the banks lends the RBI for short term purposes, as on July
31st 2019 is 5.50%. Both Repo rates and the reverse repo rates determine the MCLR. The MCLR, marginal cost of Lending Rate, the minimum rate at which
the banks can lend is set between 8.00-8.40%. All the banks add their additional cost of arranging funds and other margins to arrive at lending rates. CAGR of
the lending rates for the past 10 years is -5.5%. A drop in lending rates over the years cheaper access to funds for investment purposes which is good for the
economy.
Year BSE Sensex Lending Rates Correlation -0.80

2016-17 27338.22 7.98 Significance of correlation, t-test (LTT)
2014-15 26556.53 10.13 Alternate Hypothesis correlation is negative
2012-13 18202.10 9.98 T-statistic -3.73
2010-11 18605.18 8.88 A 0.01 Accept
2009-10 15585.21 13.38 A 0.05 Reject
2008-09 12365.55 14.13 A 0.1 Reject
Mean 21491.47 10.25

Standard Deviation 6305.66 2.06 The null hypothesis is accepted at 1% significance level but rejected at 5% and 10%
Coefficient of Variation 29% 20% significant levels. However, it can be deduced that correlation between BSE Sensex and
Skewness 0.36 0.97 Lending Rates is negative and strong at 5% significant levels
Kurtosis -0.870 0.316
16.00 35000.00
14.00 30000.00
12.00
25000.00
10.00
20000.00
8.00
15000.00
6.00 While BSE Sensex has skewness with low positive values which
10000.00 suggests that majority of the data set is centered around the
4.00
mean, WPI has a high positive skewness that suggests that the
2.00 5000.00 data is skewed towards the right. The coefficient of variation is low
0.00 0.00 for BSE Sensex but very high for Lending Rates
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
BSE Sensex has a negative kurtosis but Lending Rates have a
Lending Rates BSE Sensex positive kurtosis. However, both the variables have kurtosis lesser
than 3 that suggests that both the variables have lighter tails.
16.00
14.00 moderate t a good fit since the null hypothesis is rejected at 5%
12.00 levels.
10.00
8.00
y = -0.0003x + 15.843
6.00
4.00
2.00
0.00
0.00 5000.00 10000.00 15000.00 20000.00 25000.00 30000.00 35000.00
Crude Oil Prices and BSE Sensex
Crude Oil is the major global macro economic indicator. Crude Oil forms the backbone of all major manufacturing economies because all these economies
depend on transportation efficacy. 61% of the crude oil is controlled by OPEC and it is their monopoly on what largely governs the crude oil prices globally.
They show a CAGR OF -0.8% in the last 10 years. The recession of 2008 led to spike in crude prices globally. Similar high prices lasted a span of 4-5 years
BSE Crude Oil Prices

Year Correlation -0.54
Sensex ($/barrel)
2016-17 27338.22 45.33 Significance of correlation, t-test (LTT)
2012-13 18202.10 91.17 T-statistic -1.80
2011-12 17422.88 86.46 p value 0.055 Inference
2010-11 18605.18 87.04 A 0.01 Accept
2009-10 15585.21 71.21 A 0.05 Accept
2008-09 12365.55 53.48 A 0.1 Reject
Mean 21491.47 65.66

Standard Deviation 6305.66 21.12 The null hypothesis is rejected at 10% significance level but gets accepted at 1% and 5%
Coefficient of Variation 29% 32% significant levels. However, it can be approximated that correlation between BSE Sensex
Skewness 0.36 -0.09 and Crude Oil Prices is negative but not significant at 5% significant level
Kurtosis -0.870 -1.886
100 35000.00
90
30000.00
80
70 25000.00
60
20000.00
50
15000.00
40
30 10000.00 While BSE Sensex has skewness with low positive values which
20 suggests that majority of the data set is centered around the
5000.00 mean, Crude Oil Prices have a low negative skewness that suggests
10
0 0.00 that the data is centered around the mean. The coefficient of
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 variation is moderate for both BSE Sensex and Crude Oil Prices.
Crude Oil Prices ($/barrel) BSE Sensex Both the variables have negative kurtosis that suggests that both
100
90 29% of the correlation is explained and the regression line is a not
80
a good fit since the null hypothesis is not rejected at 5% levels.
70
60
50
y = -0.0018x + 104.23
40
30
20
10
0
0.00 5000.00 10000.00 15000.00 20000.00 25000.00 30000.00 35000.00
Foreign Exchange Rate and BSE Sensex
India follows a managed floating exchange rate system. It can be seen INR to USD Exchange rate has depreciated over the years which has given impetus to
the exports for the economy but tough on imports. The CAGR for foreign exchange in the last 10 years from 2007-2008 to 2017-2018 is 3.5%.
Year BSE Sensex Forex Correlation 0.88

2012-13 18202.10 58.51 T-statistic 5.26
2011-12 17422.88 53.42 p value 0.000381 Inference
2010-11 18605.18 46.45 A 0.01 Reject
2009-10 15585.21 45.65 A 0.05 Reject
2008-09 12365.55 48.36 A 0.1 Reject
Mean 21491.47 57.82

The null hypothesis is rejected at 1% significance level and can be deduced that correlation
between BSE Sensex and Forex is positive and strong at 1% level
Skewness 0.36 -0.32
Kurtosis -0.870 -1.669
80 35000.00
70 30000.00
60
25000.00
50
20000.00
40
15000.00
30
10000.00
20 While BSE Sensex has skewness with low positive values which
10 5000.00 suggests that majority of the data set is centered around the
mean, Forex has a low negative skewness that suggests that the
0 0.00
data is skewed towards the left. The coefficient of variation is
2008 2009 2011 2012 2013 2014 2015 2016 2017 2018
moderate for both BSE Sensex and Forex.
Forex BSE Sensex
80
70 y = 0.0012x + 31.56
60 good fit since the null hypothesis is ejected at 5% levels.
50
40
30
20
10
0
0.00 5000.00 10000.00 15000.00 20000.00 25000.00 30000.00 35000.00
Forex Linear (Forex)
FDI and BSE Sensex
FDI is a major monetary source for the Indian economy. In 2015, India overtook China and USA as the preferred destination for FDI. In 2018-2019, India
received the maximum ever FDI in the country at Rs 64 billion USD. Mauritius, Singapore, Netherlands, Japan and the US are its major FDI partners. Most of
the FDI is received in Infrastructure, Automobiles, Textiles and Railways. Its CAGR over the last 10 years stands at 5.1%
Year BSE Sensex FDI (USD mn) Correlation 0.75

2017-18 32396.83 37,366 Coefficient of determination 0.56
2016-17 27338.22 36,317 Significance of correlation, t-test
2015-16 26322.10 36,068 Null Hypothesis correlation between variables is 0.
2014-15 26556.53 24,748 Alternate Hypothesis correlation is positive
2013-14 20120.12 16,054 Degree of Freedom 8
2012-13 18202.10 18,286 T-statistic 3.19
2011-12 17422.88 23,473 p value 0.00644 Inference
2010-11 18605.18 14,939 A 0.01 Reject
2009-10 15585.21 22,461 A 0.05 Reject
2008-09 12365.55 22,697 A 0.1 Reject
Mean 21491.47 25240.90

The null hypothesis is rejected at 1% significance level and can be deduced that correlation
between BSE Sensex and FDI is positive and strong at 1% level
Skewness 0.36 0.50
Kurtosis -0.870 -1.297
40,000 35000.00
35,000 30000.00
30,000
25000.00
25,000
20000.00
20,000
15000.00
15,000
10000.00
10,000 While BSE Sensex has skewness with low positive values which
5,000 5000.00 suggests that majority of the data set is centered around the
mean, FDI has a high positive skewness that suggests that the data
0 0.00
2008 2009 2011 2012 2013 2014 2015 2016 2017 2018
is skewed towards the right. The coefficient of variation is low for
BSE Sensex but higher for Forex.
FDI (USD mn) BSE Sensex
40,000 56% of the correlation is explained and the regression line is a

35,000 y = 1.0027x + 3690.7 moderate a good fit since the null hypothesis is rejected at 5%
levels.
30,000
25,000
20,000
15,000
10,000
5,000
0
0.00 5000.00 10000.00 15000.00 20000.00 25000.00 30000.00 35000.00
Balance of Payments and BSE Sensex
India’s balance of payments is characterised by negative balance in current account due to import values greater than exports, with the deficit tripling in
2017-2018 (USD 48717MN) from 2016-2017 (USD 14417 MN). The capital account has grown from USD 35667 MN in 2017-2018 to 91390 MN in 2017-2018
primarily on account of loans and borrowings and not because of foreign investments. The CAGR of Balance of Payments stands at 12.1%.
Year BSE Sensex BOP (USD mn) Correlation 0.80

2017-18 32396.83 43,574 Coefficient of determination 0.64
2016-17 27338.22 21,550 Significance of correlation, t-test
2015-16 26322.10 17,905 Null Hypothesis correlation between variables is 0.
2014-15 26556.53 61,406 Alternate Hypothesis correlation is positive
2013-14 20120.12 15,508 Degree of Freedom 8
2012-13 18202.10 3,826 T-statistic 3.74
2011-12 17422.88 -12,831 p value 0.002844 Inference
2010-11 18605.18 13,050 A 0.01 Reject
2009-10 15585.21 13,441 A 0.05 Reject
2008-09 12365.55 -20,080 A 0.1 Reject
Mean 21491.47 15734.90

The null hypothesis is rejected at 1% significance level and can be deduced that
Coefficient of Variation 29% 152% correlation between BSE Sensex and Balance of Payments is positive and strong at 1%
level
Skewness 0.36 0.47
Kurtosis -0.870 0.511
70,000 35000.00
60,000
30000.00
50,000
40,000 25000.00
30,000
20000.00
20,000
15000.00 While BSE Sensex has skewness with low positive value which
10,000
0 10000.00 mean, WPI has a high positive skewness that suggests that the
2008 2009 2011 2012 2013 2014 2015 2016 2017 2018
-10,000 data is skewed towards the right. The coefficient of variation is low
5000.00
-20,000 for BSE Sensex but very high high for Balance of Payments.
-30,000 0.00
BSE Sensex has a negative kurtosis that suggests that both the
BOP (USD mn) BSE Sensex variables have lighter tails.
Balance of Payments has a positive kurtosis but lesser than 3 which

also suggests lighter tails.
70,000
60,000
50,000 y = 3.0271x - 49322 moderate good fit since the null hypothesis is rejected at 5%
40,000 levels.
30,000
20,000
10,000
0
0.00 5000.00 10000.00 15000.00 20000.00 25000.00 30000.00 35000.00
-10,000
-20,000
-30,000
Conclusion
It can be seen from the above research findings that :
• The following macro economic indicators have a positive correlation significant at levels of 5%
significance :
o NDP at Factor Prices

o CPI
o Gold Prices
o Forex
o FDI
o Balance of Payments
• The following macro economic indicators have a positive correlation without significance at
levels of 5%.
o WPI
• The following macro economic indicators have a negative correlation which is significant at
levels of 5% significance :
o Lending Rates
• The following macro economic indicators have a negative correlation which is not significant
at 5% levels of insignificance and tend to accept the null hypothesis of indifferent correlation.
o Crude Oil
Limitations
The following are the limitations of the research concluded :
o The data selection is of past 10 years. Hence, the findings cannot be generalised.
o The base year of WPI and CPI samples is 2012. Hence, it only considers time period from
2011-2012 to 2017-2018, and excludes data sets of periods earlier than 2012.
o Figures of macro economic indicators such as WPI, CPI, Forex, Crude Oil Prices, Gold Prices
and Lending Rates are periodic averages for the years taken into consideration.
The research is carried out as a project undertaken for the academic module “Macro-economics”
undertaken by faculty Dr. Chandan Sharma, during IPMX12, IIM-Lucknow, Noida Campus. Use of this
project for any other inference is not recommended.

Project Report MacroEconomics

Uploaded by

Copyright:

Available Formats

Project Report MacroEconomics

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Project Report MacroEconomics

Uploaded by

Copyright:

Available Formats

7/31/2019 Sensitivity of BSE

Statistical Tools Used:

• Correlation and Regression

Year BSE NDP FC (Rs Correlation 0.97

Year BSE Sensex CPI Correlation 0.95

Mean 24051.25 116.74

Year BSE Sensex WPI Correlation 0.70

Mean 25155.98 111.56

Year BSE Sensex Gold (Rs/10 gms) Correlation 0.70

Mean 21491.47 24586.38

Gold (Rs/10 gms) Linear (Gold (Rs/10 gms))

Year BSE Sensex Lending Rates Correlation -0.80

Mean 21491.47 10.25

BSE Crude Oil Prices

Mean 21491.47 65.66

Year BSE Sensex Forex Correlation 0.88

Mean 21491.47 57.82

Forex Linear (Forex)

Year BSE Sensex FDI (USD mn) Correlation 0.75

Mean 21491.47 25240.90

40,000 56% of the correlation is explained and the regression line is a

Year BSE Sensex BOP (USD mn) Correlation 0.80

Mean 21491.47 15734.90

Balance of Payments has a positive kurtosis but lesser than 3 which

o NDP at Factor Prices

You might also like