The Impact of Macroeconomic Indicators on Stock Prices in Vietnam : Luận văn thạc sĩ

UNIVERSITY OF ECONOMICS HO CHI MINH CITY International School of Business - HUYNH THANH DUNG THE IMPACT OF MACROECONOMIC INDICATORS ON STOCK PRICES IN VIETNAM MASTER OF BUSINESS (Honours) SUPERVISOR: Dr PHAM QUOC HUNG Ho Chi Minh City – 2013 i * ACKNOWLEDGEMENTS Firstly, I would like to express my very great appreciation to my supervisor – Dr Pham Quoc Hung – for his enthusiastic guidance, valuable and constructive suggestions during the planning and development of this thesis I would also like to thank all of my lecturers at International School of Business (ISB) – University of Economics Ho Chi Minh City (UEH) – for sharing their knowledge and experience during my master course Enthusiastic assistance provided by the ISB’s executive board and staffs was also greatly appreciated I am grateful to all my friends and classmates form MBus – ISB for their support Finally, I would like to express my special thanks to my wife for her support and encouragement throughout my study ii ABSTRACT This study investigates the impact of macroeconomic indicators on Vietnamese stock prices using monthly data spanning from January 2008 to May 2013 Macroeconomic variables used in this study are industrial production index, lending interest rate, consumer price index (as a proxy for inflation), and exchange rate The study employs the unit root test, cointegraion test developed by Johansen and Jesulius (1990) and Vector Error Correction Model (VECM) in order to examine the relationship between macroeconomic indicators and stock prices The empirical results reveal that there is a cointegrated relationship between the stock price and four selected macroeconomic variables in Vietnam, indicating the presence of long run equilibrium relationship In the long run, the industrial production and interest rate have significant positive effect on stock prices In contrast, consumer price index has significant negative impact on stock prices In the short run, stock prices are only affected by itself and the interest rate at onemonth lag The industrial production index and consumer price index have no effect on stock prices Furthermore, the findings show that the exchange rate does not influence stock prices in the short run as well as in the long run Keywords: Stock price, macroeconomic variables, cointegation, VECM iii TABLE OF CONTENT ACKNOWLEDGEMENTS i ABSTRACT ii TABLE OF CONTENT iii LIST OF FIGURES vi LIST OF TABLES vii LIST OF ABBREVIATIONS ix CHAPTER 1: INTRODUCTION 1.1 Research background 1.2 Research problems 1.3 Research objectives 1.4 Significance of the research 1.6 Research methodology and scope 1.7 Research structure CHAPTER 2: LITERATURE REVIEW 2.1 Theoretical framework 2.1.1 The top-down approach 2.1.2 The dividend valuation model 10 2.2 Relationship between industrial production and stock price 11 2.3 Relationship between interest rate and stock price 13 iv 2.4 Relationship between inflation and stock price 17 2.5 Relationship between exchange rate and stock price 20 2.6 Hypotheses summary 23 2.7 Research model 24 CHAPTER 3: RESEARCH METHODOLOGY 25 3.1 Research process 25 3.2 Measurement of variables 26 3.2.1 Dependent variable 26 3.2.2 Independent variables 26 3.3 Data collection and sample size 27 3.4 Model specification 28 3.5 Method of data analysis 28 3.5.1 Unit root test 29 3.5.2 The order of integration 31 3.5.3 Cointegration concept 31 3.5.4 Cointegration test 32 3.5.5 Vector Error Correction Model 33 CHAPTER 4: DATA ANALYSIS AND RESULTS 35 4.1 Descriptive statistics 35 4.2 Correlation analysis 37 4.3 Unit root test 38 v 4.4 Cointegration test 39 4.4.1 Optimal lag length selection 39 4.4.2 Cointegration test 40 4.5 Hypotheses testing 42 4.5.1 The long run relationship 43 4.5.2 The short run relationship 46 4.6 Diagnostic tests 51 4.6.1 Autocorrelation test 51 4.6.2 Normality test 52 4.6.3 Heteroskedasticity test 53 CHAPTER 5: CONCLUSIONS AND IMPLICATIONS 54 5.1 Conclusions 54 5.2 Implications 56 5.3 Limitations and further research 57 REFERENCES 59 APPENDICES 62 vi LIST OF FIGURES Figure 1.1 VN-Index from January 2001 to May 2013 Figure 2.1 VN-Index and Industrial Production Index 13 Figure 2.2 VN-Index and Interest Rate 17 Figure 2.3 VN-Index and CPI 20 Figure 2.4 VN-Index and Exchange Rate 23 Figure 2.5 Conceptual model 24 Figure 3.1 Research process 25 vii LIST OF TABLES Table 3.1 Description of variables 27 Table 4.1 Descriptive information 35 Table 4.2 Correlation matrix 37 Table 4.3 Result of unit root test at levels 38 Table 4.4 Result of unit root test after first differencing 38 Table 4.5 VAR lag order selection criteria 40 Table 4.6 Result of cointegration test 41 Table 4.7 Result of cointegrating vector 43 Table 4.8 Result of long run relationship 44 Table 4.9 Result of short run relationship 47 Table 4.10 Hypotheses testing summary 50 Table 4.11 Result of autocorrelation 51 Table 4.12 Result of normality test 52 Table 4.13 Result of heteroskedasticity 53 Table A1 Correlation Matrix 62 Table A2 Result of ADF test for LVN 63 Table A3 Result of ADF test for D(LVN) 63 Table A4 Result of ADF test for LIP 64 Table A5 Result of ADF test for D(LIP) 65 Table A6 Result of ADF test for LIR 66 viii Table A7 Result of ADF test for D(LIR) 67 Table A8 Result of ADF test for LCPI 67 Table A9 Result of ADF test for D(LCPI) 68 Table A10 Result of ADF test for LEXR 69 Table A11 Result of ADF test for D(LEXR) 70 Table A12 Result of optimal lag length selection 71 Table A13 Result of cointegration test 71 Table A14 Result of VECM 75 Table A15 Result of coefficients of ECM 76 Table A16 Serial correlation LM test 77 Table A17 Correlogram Q-statistics 79 Table A18 Heteroskedasticity test 80 ix LIST OF ABBREVIATIONS ADF Augmented Dickey Fuller CAPM Capital Asset Pricing Model CPI Consumer Price Index DVM Dividend Valuation Model ECM Error Correction Mechanism ECT Error Correction Term EViews Econometric Views EXR Exchange Rate GDP Gross Domestic Product GNP Gross National Product GSO General Statistic Office HNX-Index Hanoi Stock Price Index HNX Hanoi Stock Exchange HOSE Ho Chi Minh Stock Exchange IFS International Financial Statistics IMF International Monetary Fund IPI Industrial Production Index IPO Initial Public Offering IR Interest Rate OLS Ordinary Least Squared SPSS Statistical Package for the Social Sciences 66 Adjusted R-squared S.E of regression Sum squared resid Log likelihood Durbin-Watson stat 0.870856 0.054122 0.169896 92.88972 2.016391 S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter 0.150606 -2.947204 -2.843391 -2.906519 Table A6 Result of ADF test for LIR Null Hypothesis: LIR has a unit root Exogenous: None Lag Length: (Automatic based on SIC, MAXLAG=10) Augmented Dickey-Fuller test statistic Test critical values: 1% level 5% level 10% level t-Statistic Prob.* -0.269395 -2.602185 -1.946072 -1.613448 0.5851 *MacKinnon (1996) one-sided p-values Augmented Dickey-Fuller Test Equation Dependent Variable: D(LIR) Method: Least Squares Date: 11/05/13 Time: 12:47 Sample (adjusted): 2008M03 2013M05 Included observations: 63 after adjustments Variable Coefficient Std Error t-Statistic Prob LIR(-1) D(LIR(-1)) -0.000973 0.366532 0.003611 0.119080 -0.269395 3.078030 0.7885 0.0031 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood Durbin-Watson stat 0.135022 0.120842 0.074760 0.340933 75.01179 1.967159 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter -0.001456 0.079733 -2.317835 -2.249799 -2.291076 67 Table A7 Result of ADF test for D(LIR) Null Hypothesis: D(LIR) has a unit root Exogenous: None Lag Length: (Automatic based on SIC, MAXLAG=10) Augmented Dickey-Fuller test statistic Test critical values: 1% level 5% level 10% level t-Statistic Prob.* -3.765518 -2.602794 -1.946161 -1.613398 0.0003 *MacKinnon (1996) one-sided p-values Augmented Dickey-Fuller Test Equation Dependent Variable: D(LIR,2) Method: Least Squares Date: 11/05/13 Time: 12:48 Sample (adjusted): 2008M04 2013M05 Included observations: 62 after adjustments Variable Coefficient Std Error t-Statistic Prob D(LIR(-1)) D(LIR(-1),2) -0.471767 -0.253811 0.125286 0.112718 -3.765518 -2.251737 0.0004 0.0280 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood Durbin-Watson stat 0.416620 0.406897 0.064450 0.249227 83.03805 1.757060 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter -0.004488 0.083687 -2.614131 -2.545513 -2.587190 Table A8 Result of ADF test for LCPI Null Hypothesis: LCPI has a unit root Exogenous: None Lag Length: (Automatic based on SIC, MAXLAG=10) Augmented Dickey-Fuller test statistic t-Statistic Prob.* -0.937992 0.3066 68 Test critical values: 1% level 5% level 10% level -2.602185 -1.946072 -1.613448 *MacKinnon (1996) one-sided p-values Augmented Dickey-Fuller Test Equation Dependent Variable: D(LCPI) Method: Least Squares Date: 11/05/13 Time: 12:49 Sample (adjusted): 2008M03 2013M05 Included observations: 63 after adjustments Variable Coefficient Std Error t-Statistic Prob LCPI(-1) D(LCPI(-1)) -0.006910 0.590070 0.007367 0.102049 -0.937992 5.782245 0.3519 0.0000 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood Durbin-Watson stat 0.357878 0.347352 0.144005 1.264987 33.71116 2.239980 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter -0.014323 0.178254 -1.006703 -0.938667 -0.979945 Table A9 Result of ADF test for D(LCPI) Null Hypothesis: D(LCPI) has a unit root Exogenous: None Lag Length: (Automatic based on SIC, MAXLAG=10) Augmented Dickey-Fuller test statistic Test critical values: 1% level 5% level 10% level *MacKinnon (1996) one-sided p-values Augmented Dickey-Fuller Test Equation Dependent Variable: D(LCPI,2) t-Statistic Prob.* -3.997763 -2.602185 -1.946072 -1.613448 0.0001 69 Method: Least Squares Date: 11/05/13 Time: 12:50 Sample (adjusted): 2008M03 2013M05 Included observations: 63 after adjustments Variable Coefficient Std Error t-Statistic Prob D(LCPI(-1)) -0.407432 0.101915 -3.997763 0.0002 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood Durbin-Watson stat 0.204781 0.204781 0.143866 1.283233 33.26007 2.227993 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter -0.002302 0.161329 -1.024129 -0.990111 -1.010750 Table A10 Result of ADF test for LEXR Null Hypothesis: LEXR has a unit root Exogenous: None Lag Length: (Automatic based on SIC, MAXLAG=10) Augmented Dickey-Fuller test statistic Test critical values: 1% level 5% level 10% level t-Statistic Prob.* 2.240533 -2.601596 -1.945987 -1.613496 0.9936 *MacKinnon (1996) one-sided p-values Augmented Dickey-Fuller Test Equation Dependent Variable: D(LEXR) Method: Least Squares Date: 11/05/13 Time: 12:52 Sample (adjusted): 2008M02 2013M05 Included observations: 64 after adjustments Variable LEXR(-1) R-squared Coefficient Std Error t-Statistic Prob 0.000408 0.000182 2.240533 0.0286 -0.000835 Mean dependent var 0.004032 70 Adjusted R-squared S.E of regression Sum squared resid Log likelihood Durbin-Watson stat -0.000835 0.014321 0.012921 181.4366 2.116637 S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter 0.014315 -5.638644 -5.604912 -5.625355 Table A11 Result of ADF test for D(LEXR) Null Hypothesis: D(LEXR) has a unit root Exogenous: None Lag Length: (Automatic based on SIC, MAXLAG=10) Augmented Dickey-Fuller test statistic Test critical values: 1% level 5% level 10% level t-Statistic Prob.* -7.720233 -2.602185 -1.946072 -1.613448 0.0000 *MacKinnon (1996) one-sided p-values Augmented Dickey-Fuller Test Equation Dependent Variable: D(LEXR,2) Method: Least Squares Date: 11/05/13 Time: 12:52 Sample (adjusted): 2008M03 2013M05 Included observations: 63 after adjustments Variable Coefficient Std Error t-Statistic Prob D(LEXR(-1)) -0.980047 0.126945 -7.720233 0.0000 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood Durbin-Watson stat 0.490138 0.490138 0.014994 0.013939 175.7178 1.999668 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter 4.05E-05 0.020999 -5.546598 -5.512580 -5.533219 71 Table A12 Result of optimal lag length selection VAR Lag Order Selection Criteria Endogenous variables: LVNI LIP LIR LCPI LEXR Exogenous variables: C Date: 11/04/13 Time: 20:44 Sample: 2008M01 2013M05 Included observations: 60 Lag LogL LR FPE AIC SC HQ 177.7837 454.7450 487.5484 514.5508 545.4526 570.9384 NA 498.5303 53.57894 39.60353 40.17225* 28.88395 2.17e-09 4.90e-13 3.85e-13 3.76e-13 3.38e-13* 3.87e-13 -5.759456 -14.15817 -14.41828 -14.48503 -14.68175 -14.69795* -5.584928 -13.11099* -12.49846 -11.69257 -11.01665 -10.16020 -5.691188 -13.74856* -13.66734 -13.39274 -13.24813 -12.92298 * indicates lag order selected by the criterion LR: sequential modified LR test statistic (each test at 5% level) FPE: Final prediction error AIC: Akaike information criterion SC: Schwarz information criterion HQ: Hannan-Quinn information criterion Table A13 Result of cointegration test Date: 11/04/13 Time: 15:11 Sample (adjusted): 2008M04 2013M05 Included observations: 62 after adjustments Trend assumption: Linear deterministic trend (restricted) Series: LVNI LIP LIR LCPI LEXR Lags interval (in first differences): to Unrestricted Cointegration Rank Test (Trace) Hypothesized Trace No of CE(s) Eigenvalue None * At most At most At most At most 0.436307 0.381364 0.256564 0.153363 0.070266 Statistic 0.05 Critical Value Prob.** 98.53645 62.99521 33.22041 14.83909 4.517108 88.80380 63.87610 42.91525 25.87211 12.51798 0.0083 0.0592 0.3255 0.5879 0.6666 72 Trace test indicates cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values Unrestricted Cointegration Rank Test (Maximum Eigenvalue) Hypothesized Max-Eigen No of CE(s) Eigenvalue None At most At most At most At most 0.436307 0.381364 0.256564 0.153363 0.070266 Statistic 0.05 Critical Value Prob.** 35.54124 29.77480 18.38132 10.32199 4.517108 38.33101 32.11832 25.82321 19.38704 12.51798 0.1010 0.0941 0.3487 0.5844 0.6666 Max-eigenvalue test indicates no cointegration at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values Unrestricted Cointegrating Coefficients (normalized by b'*S11*b=I): LVNI LIP LIR LCPI LEXR -1.350601 -3.678790 -6.210863 7.537908 -0.316808 49.32717 -9.983509 20.79264 -4.244209 -2.554425 -7.571389 -5.303911 1.945530 -7.584205 0.547914 1.967939 1.157293 -2.541250 1.019373 0.292519 -1.188850 8.950211 -4.449226 31.07816 -35.70100 @TREND (08M02) 0.066231 -0.034520 0.044181 -0.167983 0.147924 -0.011572 0.002942 0.018512 0.017555 0.000923 -0.002616 0.004963 -0.001178 -0.008033 0.003026 Unrestricted Adjustment Coefficients (alpha): D(LVNI) D(LIP) D(LIR) D(LCPI) D(LEXR) -0.001745 -0.029100 0.009236 -0.014595 0.002742 Cointegrating Equation(s): 0.040250 0.008112 0.017459 -0.054098 -0.002080 0.025215 -0.003041 0.007471 0.041171 0.001839 Log likelihood 488.8995 Normalized cointegrating coefficients (standard error in parentheses) LVNI LIP LIR LCPI LEXR @TREND 73 1.000000 -36.52237 (6.43494) 5.605939 -1.457083 (1.42479) (0.42804) 0.880238 (6.69026) (08M02) -0.049038 (0.03894) Adjustment coefficients (standard error in parentheses) D(LVNI) 0.002357 (0.01676) D(LIP) 0.039303 (0.00790) D(LIR) -0.012474 (0.01111) D(LCPI) 0.019712 (0.02526) D(LEXR) -0.003703 (0.00249) Cointegrating Equation(s): Log likelihood 503.7869 Normalized cointegrating coefficients (standard error in parentheses) LVNI 1.000000 LIP 0.000000 0.000000 1.000000 LIR LCPI LEXR 1.729773 -0.393607 -2.203765 (0.45519) (0.15112) (2.57366) -0.106131 0.029118 -0.084441 (0.03224) (0.01070) (0.18228) Adjustment coefficients (standard error in parentheses) D(LVNI) -0.145714 -0.487917 (0.04322) (0.55499) D(LIP) 0.009462 -1.516417 (0.02247) (0.28857) D(LIR) -0.076704 0.281272 (0.03074) (0.39478) D(LCPI) 0.218729 -0.179853 (0.06687) (0.85873) D(LEXR) 0.003947 0.156000 (0.00714) (0.09164) Cointegrating Equation(s): Log likelihood 512.9776 @TREND (08M02) 0.005343 (0.01409) 0.001489 (0.00100) 74 Normalized cointegrating coefficients (standard error in parentheses) LVNI 1.000000 LIP 0.000000 LIR 0.000000 0.000000 1.000000 0.000000 0.000000 0.000000 1.000000 LCPI LEXR 0.255693 -0.301535 (0.07541) (1.72985) -0.010720 -0.201154 (0.00968) (0.22198) -0.375367 -1.099699 (0.05774) (1.32459) @TREND (08M02) -4.61E-05 (0.00972) 0.001820 (0.00125) 0.003115 (0.00744) Adjustment coefficients (standard error in parentheses) D(LVNI) -0.302319 0.036363 -0.151213 (0.07663) (0.56823) (0.09858) D(LIP) 0.028352 -1.579657 0.171389 (0.04199) (0.31135) (0.05401) D(LIR) -0.123103 0.436608 -0.147997 (0.05708) (0.42325) (0.07343) D(LCPI) -0.036976 0.676193 0.477539 (0.11779) (0.87336) (0.15152) D(LEXR) -0.007477 0.194247 -0.006150 (0.01323) (0.09813) (0.01702) Cointegrating Equation(s): Log likelihood 518.1386 Normalized cointegrating coefficients (standard error in parentheses) LVNI 1.000000 LIP 0.000000 LIR 0.000000 LCPI 0.000000 0.000000 1.000000 0.000000 0.000000 0.000000 0.000000 1.000000 0.000000 0.000000 0.000000 0.000000 1.000000 LEXR 1.323773 (1.19531) -0.269293 (0.16890) -3.485715 (1.66019) -6.356490 (5.28810) Adjustment coefficients (standard error in parentheses) D(LVNI) -0.389546 0.085476 -0.063451 -0.032726 (0.10846) (0.56290) (0.12485) (0.03674) D(LIP) 0.050526 -1.592142 0.149079 -0.037152 @TREND (08M02) -0.009216 (0.00658) 0.002204 (0.00093) 0.016576 (0.00914) 0.035861 (0.02911) 75 D(LIR) D(LCPI) D(LEXR) (0.06001) 0.016441 (0.07702) 0.095352 (0.16676) -0.000516 (0.01892) (0.31147) (0.06908) (0.02033) 0.358038 -0.288399 0.038267 (0.39973) (0.08866) (0.02609) 0.601686 0.344399 -0.178060 (0.86545) (0.19196) (0.05649) 0.190328 -0.013154 -0.000744 (0.09817) (0.02177) (0.00641) Table A14 Result of VECM Vector Error Correction Estimates Date: 11/04/13 Time: 14:41 Sample (adjusted): 2008M03 2013M05 Included observations: 63 after adjustments Standard errors in ( ) & t-statistics in [ ] Cointegrating Eq: CointEq1 LVNI(-1) 1.000000 LIP(-1) -0.844050 (0.28569) [-2.95446] LIR(-1) -0.416630 (0.15317) [-2.71998] LCPI(-1) 0.209796 (0.05009) [ 4.18849] LEXR(-1) -0.155164 (0.14043) [-1.10489] Error Correction: D(LVNI) CointEq1 -0.383969 (0.08796) [-4.36527] 0.034116 0.209295 0.215659 0.009854 (0.06431) (0.07376) (0.15010) (0.01616) [ 0.53050] [ 2.83756] [ 1.43679] [ 0.60978] D(LVNI(-1)) 0.255146 (0.11050) [ 2.30908] 0.020598 0.077437 -0.027697 0.001520 (0.08079) (0.09266) (0.18856) (0.02030) [ 0.25497] [ 0.83574] [-0.14689] [ 0.07487] D(LIP(-1)) -0.063699 -0.656681 D(LIP) D(LIR) 0.148656 D(LCPI) 0.317307 D(LEXR) 0.015794 76 (0.13912) [-0.45789] (0.10171) (0.11666) (0.23739) (0.02556) [-6.45638] [ 1.27431] [ 1.33663] [ 0.61793] D(LIR(-1)) 0.329048 (0.14769) [ 2.22795] 0.110337 0.213025 0.053354 -0.022887 (0.10798) (0.12385) (0.25202) (0.02713) [ 1.02183] [ 1.72008] [ 0.21170] [-0.84348] D(LCPI(-1)) -0.049892 (0.07121) [-0.70069] 0.016215 0.042429 0.507450 0.006334 (0.05206) (0.05971) (0.12151) (0.01308) [ 0.31147] [ 0.71059] [ 4.17632] [ 0.48419] D(LEXR(-1)) 0.159408 (0.73780) [ 0.21606] -0.547555 0.633606 1.974538 0.028352 (0.53942) (0.61868) (1.25901) (0.13555) [-1.01508] [ 1.02412] [ 1.56832] [ 0.20916] R-squared Adj R-squared Sum sq resids S.E equation F-statistic Log likelihood Akaike AIC Schwarz SC Mean dependent S.D dependent 0.367387 0.311895 0.403861 0.084174 6.620501 69.67620 -2.021467 -1.817359 -0.003913 0.101473 0.472479 0.279519 0.403048 0.426205 0.216319 0.350683 0.215876 0.283979 1.176004 0.061541 0.070584 0.143637 10.21050 4.422766 7.697001 89.40674 80.76952 36.00878 -2.647833 -2.373635 -0.952660 -2.443725 -2.169527 -0.748551 -0.001340 -0.001456 -0.014323 0.081243 0.079733 0.178254 Determinant resid covariance (dof adj.) Determinant resid covariance Log likelihood Akaike information criterion Schwarz criterion -0.059491 -0.152429 0.013632 0.015465 -0.640116 176.4191 -5.410132 -5.206024 0.004136 0.014406 5.22E-13 3.17E-13 459.6414 -13.48068 -12.29005 Table A15 Result of coefficients of ECM Dependent Variable: D(LVNI) Method: Least Squares Date: 11/04/13 Time: 23:10 Sample (adjusted): 2008M03 2013M05 Included observations: 63 after adjustments D(LVNI) = C(1)*( LVNI(-1) - 0.844049770702*LIP(-1) - 0.416629848836 *LIR(-1) + 0.209795920238*LCPI(-1) - 0.155163524867*LEXR(-1) ) + C(2)*D(LVNI(-1)) + C(3)*D(LIP(-1)) + C(4)*D(LIR(-1)) + C(5)*D(LCPI( -1)) + C(6)*D(LEXR(-1)) + C(7) Coefficient Std Error t-Statistic Prob 77 C(1) C(2) C(3) C(4) C(5) C(6) C(7) R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) -0.382669 0.252048 -0.063500 0.329606 -0.053253 0.231717 -0.003781 0.368664 0.301020 0.084837 0.403046 69.73982 5.450122 0.000170 0.088737 0.111747 0.140213 0.148863 0.072457 0.774040 0.011238 -4.312413 2.255533 -0.452879 2.214159 -0.734958 0.299361 -0.336483 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.0001 0.0280 0.6524 0.0309 0.4654 0.7658 0.7378 -0.003913 0.101473 -1.991740 -1.753614 -1.898084 1.972711 Table A16 Serial correlation LM test Breusch-Godfrey Serial Correlation LM Test: F-statistic 0.437974 Prob F(2,55) 0.6476 Obs*R-squared 0.874237 Prob Chi-Square(2) 0.6459 Test Equation: Dependent Variable: RESID Method: Least Squares Date: 11/04/13 Time: 15:37 Sample: 2008M03 2013M05 Included observations: 63 Presample missing value lagged residuals set to zero Variable Coefficient Std Error t-Statistic Prob C(1) 0.142454 0.176380 0.807656 0.4228 C(2) 0.197884 0.250786 0.789057 0.4335 78 C(3) 0.092590 0.172389 0.537101 0.5934 C(4) 0.056606 0.161009 0.351573 0.7265 C(5) -0.031035 0.079522 -0.390270 0.6978 C(6) 0.009911 0.745271 0.013298 0.9894 RESID(-1) -0.380526 0.427001 -0.891161 0.3767 RESID(-2) -0.166135 0.207641 -0.800108 0.4271 R-squared Adjusted R-squared 0.013877 Mean dependent var -0.003421 -0.111630 S.D dependent var 0.080635 S.E of regression 0.085017 Akaike info criterion -1.973775 Sum squared resid 0.397530 Schwarz criterion -1.701631 Log likelihood 70.17393 Hannan-Quinn criter -1.866740 Durbin-Watson stat 1.927119 79 Table A17 Correlogram Q-statistics 80 Table A18 Heteroskedasticity test Heteroskedasticity Test: Breusch-Pagan-Godfrey F-statistic Obs*R-squared Scaled explained SS 1.317312 12.73388 9.543940 Prob F(10,52) Prob Chi-Square(10) Prob Chi-Square(10) 0.2462 0.2389 0.4814 Test Equation: Dependent Variable: RESID^2 Method: Least Squares Date: 11/04/13 Time: 15:38 Sample: 2008M03 2013M05 Included observations: 63 Variable Coefficient Std Error t-Statistic Prob C LVNI(-1) LIP(-1) LIR(-1) LCPI(-1) LEXR(-1) LVNI(-2) LIP(-2) LIR(-2) LCPI(-2) LEXR(-2) 0.751325 0.014889 -0.059702 -0.012872 0.010305 -0.078440 -0.006253 -0.036062 0.020322 -0.008210 0.040780 0.291379 0.014243 0.023602 0.016970 0.007964 0.080013 0.011946 0.024370 0.016256 0.007848 0.080133 2.578516 1.045337 -2.529492 -0.758537 1.293885 -0.980344 -0.523484 -1.479752 1.250073 -1.046070 0.508901 0.0128 0.3007 0.0145 0.4516 0.2014 0.3315 0.6029 0.1450 0.2169 0.3004 0.6130 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.202125 0.048688 0.008529 0.003783 216.8023 1.317312 0.246190 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.006410 0.008744 -6.533406 -6.159208 -6.386232 2.135545