Determinants of capital structure: an emperical research of listed companies in Hose

MINISTRY OF EDUCATION & TRAINING UNIVERSITY OF ECONOMICS HO CHI MINH CITY - MAI THI PHUONG THAO DETERMINANTS OF CAPITAL STRUCTURE: AN EMPERICAL RESEARCH OF LISTED COMPANIES IN HOSE MASTER THESIS OF BUSINESS ADMINISTRATOR HO CHI MINH CITY – 2013 MINISTRY OF EDUCATION & TRAINING UNIVERSITY OF ECONOMICS HO CHI MINH CITY - MAI THI PHUONG THAO DETERMINANTS OF CAPITAL STRUCTURE: AN EMPERICAL RESEARCH IN HOSE LISTED COMPANIES Subject: Master of Business Administrator Code: 60.34.01.02 THESIS OF MASTER OF BUSINESS ADMINISTRATOR SUPERVISOR: DR VO XUAN VINH HO CHI MINH CITY – 2013 I ABSTRACT This paper attempts to investigate the determinants of the capital structure of a sample of listed companies on the Ho Chi Minh Stock Exchange from 2007 to 2012 After refining the data, this study can be able to use 271 companies to make the investigation on the determinants of the capital structure by observation in all samples, in each year and in each business sectors The relationship between PE, EPS, tangibility, profitability, size, and liquidity to the leverage (including total debt and short-term debt ratio) are gradually shown up The findings reveal that PE, tangibility, profitability, size and liquidity are highly significant However, EPS are not much effect to leverage The results confirm that in each economics status, the components of capital structure have their tendency toward the gearing Finally, the business factors also affect to the determinants of leverage will emphasis the difference in decision of capital structure in each industry II ACKNOWLEDGEMENT I would like to express my gratitude to all those who gave me the possibility to complete this thesis I want to thanks all my lecturers in course at University of Economics Ho Chi Minh City, who have empowered me with considerably useful knowledge during the time I studied, especially Dr Vo Xuan Vinh, who support in this thesis, thanks for his patience, motivation, enthusiasm, and immense knowledge to judge and comment on the contents of the subject Besides my advisor, I would like to thank the rest of my friend in eMBA class, for kindly helping me during my study and thesis processing Last but not the least; I would like to thank my family for supporting me spiritually throughout my life Although I has tried the best to complete the thesis, but errors could not be comprehensively avoided Therefore, I am also looking forward to receiving the inputs and comments from respectful lecturers and friends, so that the thesis could be extended and improved Mai Thi Phuong Thao Ho Chi Minh, November 2013 III COMMITMENT I would like to commit that this thesis, “DETERMINANTS OF CAPITAL STRUCTURE: A EMPERICAL RESEARCH OF LISTED COMPANIES IN HOSE”, was accomplished based on my individual study and research The data was collected based on the secure sources Mai Thi Phuong Thao Table of Contents TABLE OF CONTENTS CHAPTER INTRODUCTION 1.1 Background 1.2 Problem Statement 1.3 Purpose of Research 1.4 Organisation of the Study CHAPTER LITERATURE REVIEW AND HYPOTHESIS 2.1 Definition: 2.2 Theories 2.2.1 Modigliani and Miller 10 2.2.2 Trade-off theory and Pecking order theory 13 2.2.3 Agency cost models 17 2.2.4 Other models 19 2.3 Empirical study on Capital Structure in Vietnam 24 CHAPTER METHODOLOGY 26 3.1 Data Collection 27 3.2 Developing Hypotheses 27 3.2.1 Dependent Variables 28 3.2.2 Independent Variables: 30 3.2.3 Hypotheses: 30 a PE and EPS: 30 b Tangibility (TANG) 31 c Profitability (PROFIT) 32 d Firm Size (SIZE) 33 e Liquidity (LIQD) 35 3.3 Methods of Analysis 37 3.3.1 Descriptive analysis 38 3.3.2 Pearson correlation 38 3.3.3 Multiple regression analysis 38 CHAPTER DATA ANALYSIS AND FINDING 41 4.1 Descriptive statistic 41 4.2 Correlations between Variables 44 4.3 Multiple Regression Analysis 45 4.4 Testing on Regression Result 48 4.5 Conclusion 49 4.5.1 EPS with leverage 50 4.5.2 PE with leverage 51 4.5.3 Tangibility to leverage 52 4.5.4 Profit to leverage 53 4.5.5 Size to leverage 54 4.5.6 Liquidity to leverage 56 4.5.7 Determinants of capital structure from 2007 to 2012 57 4.5.8 Determinants of capital structure in observing in business sectors 59 CHAPTER IMPLICATION 61 5.1 Summary of finding and discussion 61 5.2 Limitation of thesis 64 5.3 Recommendation for future study 64 REFERENCES 66 CHAPTER 1.1 INTRODUCTION Background Capital structure decisions have been agued amongst theorists and practitioners in finance for many years The underlying question of such research is how and why companies come to the debt-equity ratios in their decision for capital structures and which determinants would affect their decision There are varying ways to define the debt ratio Some of companies prefer self financing while the others want to utilize the leverage To see how much a company relies on debt financing, the comparison between two companies is the good example: the cash-rich Microsoft (MSFT), and the hugely leveraged Amazon (AMZN) Microsoft, in 2000, claimed earnings before interest, taxes, depreciation and amortization, or EBITDA, of $11.8 billion, had a negative cash flow of $340.7 million in the same year That very low ratio reflects that Microsoft has zero long-term debt, and its short-term debts are relative to its massive assets In comparison with ultrasolvent Microsoft, Amazon looks positively deficit The extremely high debt ratio (2,723.6 divided by 1,852, or 1,470.2) reflects that its total debts significantly outstrip its total assets (Swanson, 2001) This example could show that there are differences of the decision in defining the appropriate capital structure The reason could be based on the difference in business activity or some determinants could affect to the debt ratio For a long time it has been believed that an optimal debt-equity choice exists for any firm, and that this optimal capital structure is a tradeoff between the advantages of debt financing and the disadvantages of bankruptcy risks From a firm’s perspective, debt is often a cheaper source of finance than equity because of tax advantages to be gained Debt is preferred over equity, especially where the firm does not face financial distress Since the publication of Modigliani and Miller’s (1958; 1963) seminal article, this argument had been developed by many theories try to explain variation in debt ratio across the firm Until now, the analysts don’t argue about which theories are the best use for company but they start to find which factor that will affect to the financial decision and performance of the company 1.2 Problem Statement Modigliani and Miller (1958) stated their famous irrelevance theory, where under perfect conditions, the choice of debt or equity is irrelevant When other research irrelevant, such as Myers and Majluf (1984) run the test, which is result of agency cost, as the underlying theory of how a firm comes to decide the debt-equity distribution Many other theories have been proposed and tested, but the Tradeoff Theory, including agency costs as part of the tradeoff, is still often applied and discussed in literature It seems that no perfect theory exists, and many theories explain only a part of the story Perhaps one theory cannot be sufficient for one firm’s capital structure determination (S.C Myers, 2001) In many years later, through the empirical work among the world and each country, financial economics has yet to provide agreement about which factors affect the selection of a specific leverage position The empirical evidence on capital structure in developed countries found that the choice of debt-equity ratio can be modeled subject to the agency cost (Marsh, 1982; Titman & Wessels, 1988) With the similar method, data from industrialized countries help explaining the differences in firms’ capital structure (Rajan & Zingales, 1995) YEAR=2010 Periods included: Cross-sections included: 271 Total panel (balanced) observations: 271 Variable C PE TANG PROFIT SIZE LIQD R-squared Adjusted Rsquared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) Coefficient Std Error -0.026696 6.29E-05 -2.68E-01 -0.013953 0.109429 -0.022111 0.072452 7.32E-05 5.36E-02 0.012376 0.012656 0.003821 t-Statistic Prob -0.368465 0.859862 -5.005418 -1.127467 8.646387 -5.787279 0.7128 0.3906 0.2606 0 0.334513 Mean dependent var 0.441247 0.321957 0.198639 10.4562 56.50985 26.64095 S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.241232 -0.372766 -0.293015 -0.340745 Dependent Variable: DEBT Method: Panel Least Squares Date: 11/12/13 Time: 00:33 Sample: 2007 2012 IF YEAR = 2011 Periods included: Cross-sections included: 271 Total panel (balanced) observations: 271 Variable C PE TANG PROFIT SIZE LIQD R-squared Adjusted Rsquared Coefficient Std Error 0.004219 7.48E-05 -3.22E-01 -0.020401 0.123448 -0.064931 0.068528 4.82E-05 5.12E-02 0.013377 0.011906 0.006657 t-Statistic 0.061568 1.552131 -6.291113 -1.525039 10.36816 -9.754325 Prob 0.951 0.1218 0.1284 0 0.436908 Mean dependent var 0.462528 0.426284 S.D dependent var 0.247325 S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.187334 9.299887 72.38945 41.12322 Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat -0.489959 -0.410207 -0.457938 Dependent Variable: DEBT Method: Panel Least Squares Date: 11/12/13 Time: 00:33 Sample: 2007 2012 IF YEAR=2012 Periods included: Cross-sections included: 271 Total panel (balanced) observations: 271 Variable C PE TANG PROFIT SIZE LIQD R-squared Adjusted Rsquared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) Coefficient Std Error -0.036066 0.0055% 1.7804% 4.3604% 10.3652% -5.6708% 0.084399 5.48E-05 5.88E-02 0.01923 0.014275 0.007303 t-Statistic -0.427324 0.99506 0.302842 2.26752 7.261344 -7.765012 Prob 0.6695 0.3206 0.7622 0.0242 0 0.322968 Mean dependent var 0.471153 0.310194 0.199349 10.53106 55.54316 25.28288 S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.240021 -0.365632 -0.28588 -0.333611 PANEL 2: DEPENDENT VARIABLES (STDTA) Dependent Variable: STDTA Method: Panel Least Squares Date: 11/12/13 Time: 00:28 Sample: 2007 2012 IF YEAR = 2007 Periods included: Cross-sections included: 271 Total panel (balanced) observations: 271 Variable Coefficient C PE TANG PROFIT SIZE LIQD 0.94% 0.01% -13.11% -36.17% 3.95% -0.82% R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) Std Error 0.009315 0.000311 4.45E-02 0.070112 0.003209 0.003015 0.402429 0.391154 0.102335 2.775184 236.2476 35.69237 t-Statistic Prob 1.005064 0.309374 -2.946068 -5.158686 12.32157 -2.731618 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.3158 0.7573 0.0035 0 0.0067 0.072785 0.13115 -1.699244 -1.619492 -1.667223 Dependent Variable: STDTA Method: Panel Least Squares Date: 11/12/13 Time: 00:30 Sample: 2007 2012 IF YEAR = 2008 Periods included: Cross-sections included: 271 Total panel (balanced) observations: 271 Variable Coefficient C PE TANG PROFIT SIZE LIQD 0.00583 3.09E-06 -1.17E-01 -0.037929 0.037986 -0.004756 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.181031 0.165578 0.156811 6.516301 120.5862 11.71549 Std Error 0.024335 0.000136 4.68E-02 0.016346 0.005404 0.001297 t-Statistic 0.239572 0.022749 -2.503687 -2.320321 7.029335 -3.666192 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat Prob 0.8108 0.9819 0.0129 0.0211 0.0003 0.133684 0.171666 -0.845654 -0.765903 -0.813633 Dependent Variable: STDTA Method: Panel Least Squares Date: 11/12/13 Time: 00:31 Sample: 2007 2012 IF YEAR = 2009 Periods included: Cross-sections included: 271 Total panel (balanced) observations: 271 Variable Coefficient C PE TANG PROFIT SIZE LIQD 0.010218 -8.79E-06 -2.66E-01 -0.000274 0.049767 -0.009106 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) Std Error 0.032359 0.000378 4.19E-02 0.000898 0.006555 0.002161 0.234269 0.219822 0.149781 5.945135 133.0161 16.21494 t-Statistic Prob 0.315774 -0.023271 -6.363516 -0.304668 7.59239 -4.213721 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.7524 0.9815 0.7609 0 0.152568 0.169575 -0.937388 -0.857636 -0.905367 Dependent Variable: STDTA Method: Panel Least Squares Date: 11/12/13 Time: 00:32 Sample: 2007 2012 IF YEAR = 2010 Periods included: Cross-sections included: 271 Total panel (balanced) observations: 271 Variable Coefficient C PE TANG PROFIT SIZE LIQD -0.002155 -4.75E-05 -2.47E-01 0.001553 0.05233 -0.013046 Std Error 0.053667 5.42E-05 3.97E-02 0.009167 0.009375 0.00283 t-Statistic -0.040156 -0.875633 -6.229767 0.16946 5.582094 -4.609786 Prob 0.968 0.382 0.8656 0 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.240026 0.225686 0.147136 5.737004 137.8448 16.73919 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.166567 0.16721 -0.973024 -0.893272 -0.941003 Dependent Variable: STDTA Method: Panel Least Squares Date: 11/12/13 Time: 00:33 Sample: 2007 2012 IF YEAR=2011 Periods included: Cross-sections included: 271 Total panel (balanced) observations: 271 Variable Coefficient C PE TANG PROFIT SIZE LIQD 0.049473 -8.49E-06 -2.99E-01 -0.004029 0.058959 -0.045524 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) Std Error 0.05189 3.65E-05 3.87E-02 0.010129 0.009016 0.00504 0.354779 0.342605 0.14185 5.332159 147.7608 29.14235 t-Statistic Prob 0.953426 -0.232566 -7.72088 -0.397794 6.539606 -9.031686 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.3412 0.8163 0.6911 0 0.179192 0.174951 -1.046205 -0.966453 -1.014184 Dependent Variable: STDTA Method: Panel Least Squares Date: 11/12/13 Time: 00:34 Sample: 2007 2012 IF YEAR=2012 Periods included: Cross-sections included: 271 Total panel (balanced) observations: 271 Variable Coefficient Std Error t-Statistic Prob C PE TANG PROFIT SIZE LIQD R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) Table 0.061586 0.0072% -18.3617% 4.6110% 4.3211% -4.4031% 0.239992 0.225652 0.155375 6.397431 123.0808 16.73609 0.065781 4.27E-05 4.58E-02 0.014988 0.011126 0.005692 0.936227 1.674363 -4.007199 3.076463 3.883851 -7.735475 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.35 0.0952 0.0001 0.0023 0.0001 0.183673 0.176568 -0.864065 -0.784313 -0.832044 Regression Analysis – Based on business sectors Dependent Variable: DEBT Method: Panel Least Squares Date: 11/12/13 Time: 00:34 Sample: 2007 2012 IF MAT =1 Periods included: Cross-sections included: 40 Total panel (balanced) observations: 240 Variable Coefficient Std Error t-Statistic C PE TANG PROFIT SIZE LIQD 0.001757 -7.01E-05 -2.27E-01 -0.52106 0.102679 -0.01136 0.032397 0.000216 7.57E-02 0.084841 0.007082 0.002028 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.555672 0.546177 0.187323 8.210998 64.47449 58.52749 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat Dependent Variable: DEBT 0.05423 -0.324816 -2.995103 -6.141685 14.49798 -5.602032 Prob 0.9568 0.7456 0.003 0 0.368293 0.278065 -0.487287 -0.400271 -0.452226 1.23379 Method: Panel Least Squares Date: 11/12/13 Time: 00:35 Sample: 2007 2012 IF CHEM=1 Periods included: Cross-sections included: 14 Total panel (balanced) observations: 84 Variable C PE TANG PROFIT SIZE LIQD R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) Coefficient 0.044204 -0.00327 -2.98E-01 -0.539319 0.108103 -0.029874 0.586787 0.560299 0.177077 2.445781 29.34016 22.1529 Std Error t-Statistic 0.068184 0.001452 1.15E-01 0.216248 0.012651 0.005712 Prob 0.648299 -2.25112 -2.59407 -2.49399 8.54505 -5.22976 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.5187 0.0272 0.0113 0.0147 0 0.353838 0.267044 -0.55572 -0.38209 -0.48592 0.977197 Dependent Variable: DEBT Method: Panel Least Squares Date: 11/12/13 Time: 00:36 Sample: 2007 2012 IF CGOODS=1 Periods included: Cross-sections included: 13 Total panel (balanced) observations: 78 Variable Coefficient C PE TANG PROFIT SIZE LIQD 0.055433 7.04E-05 -7.17E-02 -1.00983 0.105321 -0.04506 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood 0.727717 0.708809 0.102012 0.749271 70.49195 Std Error 0.058448 8.18E-05 7.85E-02 0.21542 0.010512 0.007138 t-Statistic 0.948406 0.860204 -0.91413 -4.68772 10.0194 -6.31251 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Prob 0.3461 0.3925 0.3637 0 0.449947 0.189044 -1.65364 -1.47235 -1.58107 F-statistic Prob(F-statistic) 38.48618 Durbin-Watson stat 0.946966 Dependent Variable: DEBT Method: Panel Least Squares Date: 11/12/13 Time: 00:36 Sample: 2007 2012 IF CONS=1 Periods included: Cross-sections included: 51 Total panel (balanced) observations: 306 Variable Coefficient C PE TANG PROFIT SIZE LIQD -0.01439 0.000249 -1.62E-01 -0.03516 0.124701 -0.06513 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.60643 0.599871 0.195728 11.49288 67.92897 92.45075 Std Error 0.027185 0.000381 5.38E-02 0.013427 0.006398 0.007835 t-Statistic Prob -0.52923 0.652414 -3.00703 -2.61851 19.48946 -8.31274 0.597 0.5146 0.0029 0.0093 0 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.447102 0.309424 -0.40477 -0.33175 -0.37557 0.877952 Dependent Variable: DEBT Method: Panel Least Squares Date: 11/12/13 Time: 00:39 Sample: 2007 2012 IF EWGO=1 Periods included: Cross-sections included: 21 Total panel (balanced) observations: 126 Variable C PE TANG PROFIT SIZE LIQD Coefficient 0.036403 0.000333 4.54E-02 -0.02822 0.058575 -0.00404 Std Error 0.045489 0.000182 7.42E-02 0.009095 0.010402 0.001956 t-Statistic 0.800259 1.827085 0.612304 -3.10325 5.631027 -2.06312 Prob 0.4251 0.0702 0.5415 0.0024 0.0413 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.396576 0.371434 0.193797 4.506854 31.04675 15.77305 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.339143 0.244439 -0.39757 -0.26251 -0.3427 0.516069 Dependent Variable: DEBT Method: Panel Least Squares Date: 11/12/13 Time: 00:40 Sample: 2007 2012 IF FOOD=1 Periods included: Cross-sections included: 38 Total panel (balanced) observations: 228 Variable Coefficient C PE TANG PROFIT SIZE LIQD 0.026964 0.001618 -3.51E-01 -0.36527 0.098352 -0.02454 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.409216 0.39591 0.206107 9.430579 39.61626 30.75431 Std Error 0.041581 0.001143 9.19E-02 0.170462 0.009293 0.004506 t-Statistic Prob 0.648462 1.41523 -3.81326 -2.14283 10.58347 -5.44552 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.5174 0.1584 0.0002 0.0332 0 0.375046 0.265181 -0.29488 -0.20463 -0.25847 0.598807 Dependent Variable: DEBT Method: Panel Least Squares Date: 11/12/13 Time: 00:40 Sample: 2007 2012 IF IND=1 Periods included: Cross-sections included: 34 Total panel (balanced) observations: 204 Variable C PE TANG Coefficient 0.041758 2.04E-05 -8.67E-02 Std Error 0.04301 0.000554 6.82E-02 t-Statistic 0.970906 0.036869 -1.27153 Prob 0.3328 0.9706 0.205 PROFIT SIZE LIQD -0.28466 0.089621 -0.03232 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.388366 0.37292 0.190108 7.155955 52.25441 25.14456 0.129819 0.009331 0.005735 -2.19278 9.604961 -5.63598 0.0295 0 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.379645 0.240071 -0.45348 -0.35588 -0.414 0.581575 Dependent Variable: DEBT Method: Panel Least Squares Date: 11/12/13 Time: 00:40 Sample: 2007 2012 IF TECH=1 Periods included: Cross-sections included: Total panel (balanced) observations: 48 Variable Coefficient C PE TANG PROFIT SIZE LIQD 0.010903 0.003408 -2.50E-01 -0.17349 0.086234 -0.03238 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.539208 0.484352 0.180281 1.365049 17.3312 9.82948 0.000003 Std Error t-Statistic 0.078788 0.003382 1.49E-01 0.07224 0.015962 0.009482 Prob 0.13838 1.007485 -1.67651 -2.40156 5.402336 -3.4146 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.357704 0.251057 -0.47213 -0.23823 -0.38374 1.822236 Dependent Variable: DEBT Method: Panel Least Squares Date: 11/12/13 Time: 00:41 Sample: 2007 2012 IF REAL=1 Periods included: Cross-sections included: 26 Total panel (balanced) observations: 156 Variable Coefficient Std Error t-Statistic 0.8906 0.3195 0.1011 0.0208 0.0014 Prob C PE TANG PROFIT SIZE LIQD 0.008932 1.52E-05 -1.16E-01 0.001097 0.096325 -0.02985 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.495721 0.478912 0.199974 5.99843 32.79753 29.49085 0.040171 6.48E-05 7.61E-02 0.001614 0.008595 0.008943 0.222355 0.235272 -1.52908 0.679967 11.20668 -3.3383 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.8243 0.8143 0.1284 0.4976 0.0011 0.41376 0.277024 -0.34356 -0.22626 -0.29592 0.917449 Dependent Variable: DEBT Method: Panel Least Squares Date: 11/12/13 Time: 00:41 Sample: 2007 2012 IF TRANS=1 Periods included: Cross-sections included: 17 Total panel (balanced) observations: 102 Variable Coefficient C PE TANG PROFIT SIZE LIQD -0.03468 0.000199 1.81E-01 0.123124 0.07598 -0.02394 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.609113 0.588755 0.183367 3.227854 31.37919 29.91909 Dependent Variable: DEBT Method: Panel Least Squares Date: 11/12/13 Time: 00:41 Std Error 0.04658 0.000147 1.07E-01 0.141684 0.016236 0.007416 t-Statistic -0.74445 1.350266 1.700778 0.869004 4.679709 -3.22769 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat Prob 0.4584 0.1801 0.0922 0.387 0.0017 0.410594 0.285937 -0.49763 -0.34322 -0.43511 0.601503 Sample: 2007 2012 IF OT=1 Periods included: Cross-sections included: Total panel (balanced) observations: 54 Variable Coefficient Std Error t-Statistic 0.070939 3.65E-05 1.00E-01 0.228787 0.014033 0.006257 Prob 0.897198 1.263059 -3.88255 -1.93908 6.964806 -3.5391 0.3741 0.2127 0.0003 0.0584 0.0009 C PE TANG PROFIT SIZE LIQD 0.063646 4.61E-05 -3.88E-01 -0.44364 0.097734 -0.02214 R-squared Adjusted Rsquared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.650583 Mean dependent var 0.317303 0.614186 0.165837 1.320089 23.58201 17.87437 S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.266988 -0.65119 -0.43019 -0.56596 1.085882 Table Wald Test – Joint Significant effect Equation: EQ01 Test Statistic Value F-statistic Chi-square df Probability 1030.272 (6, 1619) 6181.63 0 Null Hypothesis Summary: Normalized Restriction (= 0) C(1) C(2) C(3) C(4) C(5) C(6) Restrictions are linear in Value Std Err -5.36E-05 6.13E-05 -4.96E-06 -0.153097 -0.001237 0.095788 0.014233 3.37E-05 1.53E-06 0.023686 0.001172 0.003028 coefficients Equation: EQ02 Test Statistic Value F-statistic Chi-square df Probability 332.4337 (6, 1619) 1994.602 0 Null Hypothesis Summary: Normalized Restriction (= 0) Value C(1) C(2) C(3) C(4) C(5) C(6) Std Err 0.012593 6.87E-07 -8.00E-06 -0.203148 -0.000483 0.046725 0.010183 2.41E-05 1.10E-06 0.016946 0.000838 0.002166 Restrictions are linear in coefficients Table 10 Redundant Variables Test a Redundant Variables: PE F-statistic Log likelihood ratio Prob 3.30883 F(1,1619) Prob Chi3.319745 Square(1) Test Equation: Dependent Variable: DEBT Method: Panel Least Squares Date: 11/26/13 Time: 18:54 Sample: 2007 2012 Periods included: Cross-sections included: 271 Total panel (balanced) observations: 1626 0.0691 0.0685 Variable Coefficient Std Error t-Statistic C EPS TANG PROFIT SIZE LIQD -3.83E-05 -5.17E-06 -0.1544 -0.00125 0.096228 -0.01704 0.014244 1.53E-06 0.023692 0.001173 0.00302 0.001251 -0.002692 -3.384112 -6.516935 -1.064013 31.86225 -13.62595 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.430106 0.428347 0.206094 68.80886 263.9556 244.5267 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat Prob 0.9979 0.0007 0.2875 0 0.391416 0.272583 -0.317289 -0.297385 -0.309904 0.722831 b Redundant Variables: PROFIT F-statistic Log likelihood ratio Prob 1.113626 F(1,1619) Prob Chi1.118057 Square(1) 0.2915 0.2903 Test Equation: Dependent Variable: DEBT Method: Panel Least Squares Date: 11/26/13 Time: 18:56 Sample: 2007 2012 Periods included: Cross-sections included: 271 Total panel (balanced) observations: 1626 Variable C PE EPS TANG SIZE LIQD Coefficient Std Error 1.15E-04 6.15E-05 -4.9E-06 -0.15309 0.095784 -0.0172 0.014233 3.37E-05 0.00000153 0.023687 0.003028 0.001237 t-Statistic 0.008107 1.824474 -3.229021 -6.462924 31.63538 -13.90983 Prob 0.9935 0.0683 0.0013 0 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.430877 0.429121 0.205954 68.71576 265.0565 245.297 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.391416 0.272583 -0.318643 -0.298739 -0.311258 0.719288 c Redundant Variables: EPS F-statistic Log likelihood ratio 10.48264 Prob F(1,1619) Prob Chi10.49403 Square(1) 0.0012 0.0012 Test Equation: Dependent Variable: DEBT Method: Panel Least Squares Date: 11/26/13 Time: 18:56 Sample: 2007 2012 Periods included: Cross-sections included: 271 Total panel (balanced) observations: 1626 Variable Coefficient Std Error t-Statistic C PE TANG PROFIT SIZE LIQD -3.51E-03 6.96E-05 -0.14587 -0.00121 0.093916 -0.01703 0.014235 3.37E-05 0.023649 0.001175 0.002981 0.001253 -0.2466 2.066332 -6.16817 -1.02574 31.50918 -13.5873 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.427586 0.425819 0.206549 69.11313 260.3685 242.0237 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat Prob 0.8053 0.039 0.3052 0 0.391416 0.272583 -0.31288 -0.29297 -0.30549 0.710522