Professional Documents
Culture Documents
SCHOOL OF BUSINESS
FINAL PROJECT
Team Members:
I. TABLE OF CONTENTS
I. Table of contents...
II. Acknowledgement.......
III. Table of members contribution...
IV. Introduction..
V. Data.
VI. Methodology..
VII. Result and Interpretations
VIII. Conclusion...
IX. References...
II. ACKNOWLEDGEMENT
We would like to express my special thanks of gratitude to our lecturer Dr. Nguyen Phuong Anh
who gave us the opportunity to do this project on the topic Using Eviews for Capital Asset
Pricing Model (CAPM). We appreciate the chance that we were provided with wholehearted
support, caring guidance, and continuous encouragement throughout the course and through
the process of making this project. This accomplishment would not have been possible without
Dr. Nguyen Phuong Anh.
III. INTRODUCTION
The objective of this final project is to conduct an optimal portfolio of five stocks in order to
investigate whether CAPM can explain the returns of some companies in Construction Real
Estate Industry. Moreover, if the explaination is feasible, how its sensitivity to the market is.
All the data is based on historical data of five stocks. This reflects how stocks price is sensitive
to market changes for conducting expected return of the portfolio. We use Eviews (version 6.0)
and CAPM test for the Beta test, volatility of the stocks when compared with the market. We
collected monthly prices of not only these following stocks: Vingroup (VIC), Hung Long Joint
Stock Company (KHL), Coteccons Joint Stock Company (CTD), Nam Long Group
Corporation (NLG), Kinh Bac City Group (KBC) but also VN-Index. The data are collected
from http://www.cophieu68.com and http://investment.com
IV. DATA
The data below was collected from http://www.cophieu68.com and http://investment.com/ from
May 2013 to April 2016
Date
(YYYYMMDD) VNINDEX T-BILL CTD KBC KHL NLG VIC
20160429 598.4 6.934 174 13.8 1.9 23.4 52.5
20160331 561.2 6.868 188 12.7 2.5 22.7 47.2
20160229 559.4 6.863 153 13.1 1.8 22.6 44.1
20160129 545.2 7.037 137 11.8 1.4 23.4 47.6
20151231 579 7.052 153 13.1 1.5 22.3 45.7
20151130 573.2 7.06 148 13 1.5 22 42.8
20151030 607.4 7.062 125 14.4 1.3 20.9 45.5
20150930 562.6 7.041 98.5 12.6 1.4 20 41.7
20150831 564.8 7.002 97 12.2 1.8 19.1 41.5
20150731 621.1 6.776 86 15.9 1.9 18.5 43.1
20150630 593 6.785 76.5 15.5 2.5 18.3 43.5
20150529 569.6 6.538 79 15.8 2.9 18.8 37.6
20150427 562.4 6.493 77 15.7 2.7 18.7 38.5
20150331 551.1 6.476 70.5 16 3.4 16.9 37.2
20150227 592.6 6.433 60 16.3 3.7 17.2 39.2
20150130 576.1 6.645 60 15.7 3.7 16.6 38
20141231 545.6 7.15 57 15.9 3.9 17.2 37.8
20141128 566.6 6.893 61 16.7 3.8 16.9 38.3
20141030 600.8 6.302 61 16.1 3.6 16.8 37.7
20140930 598.8 6.66 66 15 3.9 16.6 38
20140829 636.6 8.038 62 14.3 4.4 17.7 46.3
20140731 596.1 8.426 62.5 10.5 4 16.6 37.8
20140630 578.1 8.577 61.5 11.2 4.1 16.7 32.7
20140530 562 8.694 62 10.1 3.4 16.6 35
20140429 578 8.465 62 11.2 3.8 19 33.2
20140331 591.6 8.495 63 13 5.1 17.6 37.5
20140228 586.5 8.67 63 12.9 4.1 17.2 39.1
20140127 556.5 8.85 66 10.7 3 16.1 39.1
20131231 504.6 8.95 50.5 9.8 3 15.6 35.8
20131129 507.8 8.9 45.3 9.5 3.6 16.5 35.2
20131031 497.4 8.85 42.7 9.4 2 18.6 34.5
20130930 492.6 8.93 37.5 8.5 1.4 21.2 31.9
20130830 472.7 8.81 35 7.6 1.4 20.3 31.9
20130731 491.9 8.93 37.2 7 1.7 21.2 32.2
20130628 481.1 8.9 36.9 7.4 2.1 22.3 31.9
20130531 518.4 8.94 37.8 8.5 2.2 16.1 34.7
20130426 474.5 9.208 34.8 6.6 2.5 31.7
V. METHODOLOGY
To conduct this test, we use the software package EViews (version 6.0) as appropriate tools to
test the correlation coefficient between every two of the chosen five stocks with VN-Index. The
data is collected monthly and entered into the file name RESP, including 36 observations in
8 columns: the first one is date (YYMMDD), the next one is prices of VN-Index, the third one
is the risk-free rate of T-bill and the five last are prices of chosen stocks (CTD, KBC, KHL,
NLG, VIC)
We consider the following situation:
CAPM test:
2. Testing for CAPM: CTD, KBC, KHL, NLG, VIC and VN_Index
a. Generating the return for each stock
b. For CTD: Genr RCTD=100*LOG(CTD/CTD(-1))
With (-1): using one period lagged observation of the series.
a. Generating the T-Bill from using annualized into using monthly data.
- For T-Bill: Genr T= T/12
b. Generating the excess return for each stock
- For VIC: Genr ERVIC= RVIC-T
- For KHL: Genr ERKHL= RKHL-T
- For NLG: Genr ERNLG= RNLG-T
- For CTD: Genr ERCTD= RCTD-T
- For KBC: Genr ERKBC= RKBC-T
- For VN- INDEX: Genr ERINDEX= RVINDEX- T
c. Plot the data
(the goal of this step mainly focuses on if the series appear to move together)
- Object\New object => graph VN_INDEXnDXG => type: ERINDEX ERDXG
- This will appear in the graph under the name INDEXnKDC, similarly to others:
ERINDEX ERHAG (graph INDEXnHAG)
ERINDEX ERQCG (graph INDEXnQCG)
GRAPH:
28
24
20
16
12
-4
-8
-12
13M07 14M01 14M07 15M01 15M07 16M01
ERVNINDEX ERCTD
20
10
-10
-20
-30
13M07 14M01 14M07 15M01 15M07 16M01
ERVNINDEX ERVIC
40
30
20
10
-10
-20
13M07 14M01 14M07 15M01 15M07 16M01
ERVNINDEX ERNLG
60
40
20
-20
-40
13M07 14M01 14M07 15M01 15M07 16M01
ERVNINDEX ERKHL
40
30
20
10
-10
-20
-30
13M07 14M01 14M07 15M01 15M07 16M01
ERVNINDEX ERKBC
CAPM1
8
Series: Residuals
7 Sample 2013M05 2016M04
Observations 36
6
Mean -3.45e-16
5 Median -1.770363
Maximum 19.80636
4 Minimum -13.07770
Std. Dev. 8.170314
3
Skewness 0.384288
Kurtosis 2.485328
2
Jarque-Bera 1.283396
1
Probability 0.526398
0
-15 -10 -5 0 5 10 15 20
CAPM4
9
Series: Residuals
8 Sample 2013M05 2016M04
Observations 36
7
6 Mean -1.48e-16
Median 0.074530
5 Maximum 13.44890
Minimum -14.95312
4
Std. Dev. 6.431619
3 Skewness -0.132863
Kurtosis 2.812315
2
Jarque-Bera 0.158754
1 Probability 0.923692
0
-15 -10 -5 0 5 10 15
According to statistics, a confidence level of 95% gives a t-statistic value equal to 1.96. A
special case of the t-test occurs if we test the hypothesis in which a parameter equals zero:
Ho: Bi=0. If this hypothesis is rejected, then we conclude that the regression has a
significant value for explaining the regression; if the hypothesis is not rejected, the
regression has no significant explanatory value. According to this rule, the t-statistics is less
than 1.96. Given our hypothesis condition that the intercept is equal to zero, if the t-
statistics is higher than 1.96, our hypothesis rule does not work here, and the CAPM model
is rejected in the portfolio. Moreover, we can see the relationship between the t-statistic
values and the P values in statistics gives us an insight into the CAPM model. In this case,
in the confidence level of 5%, when the P values are higher than 0.05, then the CAPM
model does not hold in the portfolio which means we reject the null hypothesis.
After conducting the test, we found that the excess return on the market proxy has a significant
explanatory power for the excess return of KBC and VIC stock. Durbin Watson stat of KBC
is 1.920405 and VIC is 2.636407, which indicate that there is no autocorrelation between the
error terms. The test of Bera-Jarque normality tests has p-value is 1.283396 for KBC and
0.158754 for VIC, so they follow the normal distribution. However, for the three other stocks
(CTD, KHL, NLG), the excess return on the market proxy has an insignificant explanatory
power for the excess return on the stocks.
VII. CONCLUSION:
This project tests the validity of CAPM model for the five typical stocks in HOSE. Our result
have little power to conclude that the whether CAPM model is supportive of the theorys basic
hypothesis that higher risk (beta) is associated with a higher level of return.
VIII. REFERENCES
BROOKS, C. (2008). Introductory: Econometrics for Finance. Cambridge, U.K.: Cambridge University
Press.
CHOUDHARY, K., & CHOUDARY, S. (2010). Testing Capital Asset Pricing Model: Empirical
Evidences from India Equity Market. Eurasian Journal of Business and Economics, 3(6), 127-
138.