Key Determinants of Leverage

of Listed Firms in DSE.

Group-04
 Group Information
Group no: 04
Section: C
Name Id

MD. A. K. M. MAHMUDUL HASAN 20-036

MD. SHEIKH RASHEL 27-045

MD. MUSLIMUR RAHMAN 27-060

ATIQUAL ISLAM 27-209

INSAN KAMAL SAFAT 27-189

Md. A. K. M. MAHMUDUL HASAN
ID No: 27-036
 HYPOTHESIS
 An act in statistics whereby an analyst
tests an assumption regarding a
population parameter

 Used to test a relationship between

two or more variables
 ROA (Return on Asset)
Positive impact on firm leverage

H1: The likelihood of Leverage is positively related to ROA (β1 > 0)

Investment

Positive and negative impact on leverage

H2a: The likelihood of leverage is positively related to investment (β2 > 0)

H2b: The likelihood of leverage is negatively related to investment (β2 < 0)

 Cash Holding
Leverage and corporate cash holdings have a negative relationship. Leverage and cash holdings are
substitutes.

H3: The likelihood of leverage is negatively related to cash holding (β3 < 0)

NWC (Net Working Capital)

Working capital is calculated as net total current assets, but the netted amount may not always be a
positive number. It can be zero or even negative.

H4: The likelihood of leverage is negatively related to net working capital (β4 < 0)
 Firm size
Larger firms (by sales or employment) have higher leverage. Therefore, the leverage generally
has a negative effect on firm size.

H5: The likelihood of leverage is negatively related to firm size (β5 < 0)

Firm Age
Results reveal that there is a negative and convex relationship between firm age and leverage

H6: The likelihood of leverage is negatively related to firm age (β6 < 0)

Sales Growth
Sales growth has a positive effect on firm value.

H7: The likelihood of leverage is positively related to sales growth (β7 > 0)
INSAN KAMAL SAFAT
27-189
Dependent Variable
• A dependent variable is the variable being tested and measured in a
statistical or scientific method. In the given statistical data, leverage is
the dependent variable(y).
Independent Variable

• An independent variable is the variable that changed and controlled in a

statistical experiment to test the effect on dependent variable.
Independent of this regression model will be listed here.
•1. Return on Assets (ROA)
•2. Investment of the particular firm ( Investment)
•3. Cash/Total assets ( Cash_holdin)
•4. Net Working Capital (NWC)
•5. Number of years of operation (Firm_age)
•6. Total assets of firm ( Firm_size)
•7. Sales growth.

•
Multiple Regression Model

Leverage = a + β1 ROA + β2 Investment + β3 Cash_holding + β4 NWC + β5

Firm_size + β6 Firm_age + β7 Sales_growth + e
Md. Muslimur Rahman
ID No: 27-060
 Preliminary Test

• Descriptive Statistics
• Multicollinearity Test
• Heteroscedasticity Test
• Autocorrelation Test
 Descriptive Statistics.
Mean Median Std. Dev. min max N
Leverage 7.169 .488 150.595 0 3682.936 598
ROA .128 .055 0.984 -1.798 22.742 598
Cash holding .148 .034 1.240 -.444 28.824 598
Investment .08 .037 0.914 -13.527 17.478 598
NWC .102 .193 1.426 -13.469 23.76 598
Sales growth .269 .082 1.685 -1 33.782 598
L firm size 7.616 7.303 1.989 0 13.392 598
Multicollinearity Test

Pearson Correlation matrix

Variables (1) (2) (3) (4) (5) (6) (7)
(1) ROA 1.000
(2) Investment -0.562 1.000
(3) Cash_holding 0.299 0.556 1.000
(4) NWC 0.694 -0.407 0.220 1.000
(5) Firm_Age -0.059 0.054 0.026 -0.149 1.000
(6) Sales_growth 0.054 -0.002 -0.015 -0.004 0.043 1.000
(7) L_firm_size -0.087 0.132 0.030 0.004 0.228 0.020 1.000
Source: Authors’ Estimation

Result No Multicollinearity Problem

Multicollinearity Test

Variance Inflation Factor(VIF) Test.

VIF 1/VIF
Investment 9.899 .101
ROA 7.785 .128
Cash holding 7.384 .135
NWC 2.052 .487
L firm size 1.106 .904
Firm Age 1.104 .906
Sales growth 1.037 .965
Mean VIF 4.338
Source: Authors’ Estimation
.

Result No Multicollinearity Problem

Heteroscedasticity Test
Command: xttest3 after run FE Model

Modified Wald test for groupwise heteroskedasticity

in fixed effect regression model

H0: sigma(i)^2 = sigma^2 for all i

chi2 (50) = 1.2e+05

Prob>chi2 = 0.0000

Since the p value is less than 0.05

Result there is a problem of heteroscedasticity.
 Autocorrelation Test.
Wooldridge test for autocorrelation in panel data

H0: no first-order autocorrelation

F( 1, 49) = 6.096

Prob > F = 0.0171

Since the p value is less than 0.05

Result there is a problem of Autocorrelation.
 Solution
of Autocorrelation and Heteroscedasticity Problem

 If there is ONLY heteroscedasticity problem, needs to be used "vce(robust)" option after FE and
RE models.

 If there is ONLY autocorrelation problem, needs to be used FE and RE models with AR(1)
disturbance.

 If there are BOTH heteroscedasticity and autocorrelation problems, run FE and RE

models with "Clustered Robust" option which takes both heteroscedasticity and
autocorrelation problem into consideration.
MD. SHEIKH RASHEL
27-045
 Pooled OLS regression model

Table
Table N o. : Pooled OLS model No. : Pooled OLS model
Leverage Coef. St.Err. t-value p-value [95% Conf Interval] Sig
ROA 35.783 3.088 11.59 0 29.718 41.847 ***
Investment -145.924 3.747 -38.94 0 -153.284 -138.565 ***
Cash_holding 87.97 2.386 36.87 0 83.285 92.656 ***
NWC -1.247 1.094 -1.14 .255 -3.396 .902
Firm_Age -.176 .063 -2.78 .006 -.3 -.052 ***
Sales_growth -.734 .658 -1.12 .265 -2.027 .558
L_firm_size .56 .576 0.97 .331 -.571 1.69
Constant 3.02 4.467 0.68 .499 -5.753 11.793

Mean dependent var 7.169 SD dependent var 150.595

R-squared 0.969 Number of obs 598
F-test 2648.337 Prob > F 0.000
Akaike crit. (AIC) 5629.178 Bayesian crit. (BIC) 5664.327
 FGLS Regression Model
FGLS regression model

Table N o. : Regression results

Leverage Coef. St.E rr. t-value p-value [95% Conf Interval] Sig
ROA 47.662 3.135 15.20 0 41.517 53.807 ***
Investment -128.57 3.834 -33.54 0 -136.084 -121.056 ***
Cash_holding 77.711 2.416 32.16 0 72.975 82.447 ***
NWC 2.109 1.131 1.86 .062 -.107 4.325 *
Firm_Age -.128 .123 -1.04 .299 -.37 .114
Sales_growth .096 .515 0.19 .852 -.913 1.105
L_firm_size -1.01 1.016 -0.99 .32 -3 .981
Constant 11.099 8.222 1.35 .177 -5.016 27.214

Mean dependent var 7.169 SD dependent var 150.595

Overall r-squared 0.967 Number of obs 598
Chi-square 25920.253 Prob > chi2 0.000
R-squared within 0.977 R-squared between 0.883
*** p<.01, ** p<.05, * p<.1
 Breusch and Pagan Lagrangian
Multiplier test
Breusch and Pagan Lagrangian multiplier test for random effects
Coef.
Chi-square test value 140.25
P-value 0

The p-value associated with the test statistic indicates the probability of obtaining the
observed test statistic (or a more extreme value) under the assumption of homoscedasticity.
The p-value is less than the chosen significance level (e.g., 0.05), it suggests rejecting the null
hypothesis in favor of heteroscedasticity and conclude that there is significant evidence of
heteroscedasticity
ATIQUAL ISLAM
27-209
Mean 7.169 SD 150.595
dependent dependent
var var
R-squared 0.978 Number of
Fixed
598
Effect (FE) Model
obs
F-test 275.826 Prob > F 0.000
Akaike crit. 5370.132 Bayesian
(AIC) Autocorrelation and crit. heteroscedasticity
Clustered Robust to
(BIC) solve

Leverage Coef. St.Err. t-value p-value [95% Conf Interval] Sig

ROA 45.904 20.479 2.24 .03 4.75 87.058 **

Investment -127.002 25.6 -4.96 0 -178.446 -75.558 ***

Cash_holding 76.421 15.583 4.90 0 45.106 107.735 ***

NWC 3.56 2.78 1.28 .206 -2.027 9.147

Firm_Age .437 .483 0.90 .371 -.534 1.408

Sales_growth .051 1.205 0.04 .966 -2.37 2.472

L_firm_size -7.436 3.363 -2.21 .032 -14.194 -.677 **

Constant 42.607 27.91 1.53 .133 -13.481 98.695

 Random Effect (RE) Model
Mean dependent 7.169 SD dependent var 150.595
var
Overall r-squared 0.968 Number of obs 598
Autocorrelation and heteroscedasticity Clustered Robust to
solve Chi-square 1484.631 Prob > chi2 0.000
R-squared within 0.977 R-squared 0.891
between
*** p<.01, ** p<.05, * p<.1

Leverage Coef. St.Err. t-value p-value [95% Conf Interval] Sig

ROA 43.591 20.197 2.16 .031 4.005 83.177 **

Investment -133.58 24.383 -5.48 0 -181.371 -85.789 ***

Cash_holding 80.596 14.837 5.43 0 51.515 109.677 ***

NWC 1.418 2.224 0.64 .524 -2.941 5.778

Firm_Age -.122 .135 -0.90 .366 -.387 .143

Sales_growth -.168 1.266 -0.13 .894 -2.65 2.314

L_firm_size -.96 .963 -1.00 .319 -2.847 .928

Constant 11.448 6.783 1.69 .091 -1.846 24.742 *

 Hausman Test
Excluding “vce(cluster robust)”

For Random Effects

For Fixed Effects Mean dependent var 7.169 SD dependent var 150.595

R-squared 0.978 Number of obs 598 Overall r-squared 0.968 Number of obs 598
F-test 3399.088 Prob > F 0.000 Chi-square 22593.150 Prob > chi2 0.000
Mean 5372.132 Bayesian crit. 5407.281 R-squared within 0.977 R-squared between 0.891
dependent var (BIC)
*** p<.01, ** p<.05, * p<.1 *** p<.01, ** p<.05, * p<.1

Leverage Coef. St.Err. t- p- [95% Interval] Sig

Leverage Coef. St.Err. t- p- [95% Interval] Sig value value Conf
value value Conf
ROA 43.591 3.169 13.76 0 37.38 49.802 ***
ROA 45.904 3.347 13.71 0 39.329 52.479 *** Investment -133.58 3.897 -34.28 0 -141.217 -125.943 ***
Investment -127.002 4.079 -31.13 0 -135.015 -118.989 ***
Cash_holding 80.596 2.466 32.69 0 75.763 85.428 ***
Cash_holding 76.421 2.568 29.75 0 71.375 81.466 ***
NWC 1.418 1.153 1.23 .219 -.841 3.678
NWC 3.56 1.214 2.93 .004 1.175 5.945 *** Firm_Age -.122 .105 -1.17 .243 -.327 .083
Firm_Age .437 .304 1.44 .151 -.16 1.034
Sales_growth -.168 .599 -0.28 .779 -1.342 1.005
Sales_growth .051 .585 0.09 .931 -1.099 1.201

L_firm_size -.96 .879 -1.09 .275 -2.682 .763

L_firm_size -7.436 1.819 -4.09 0 -11.008 -3.863 ***
Constant 42.607 12.459 3.42 .001 18.132 67.082 *** Constant 11.448 7 1.64 .102 -2.271 25.167
 Hausman (1978) specification test
Coef.
Chi-square test 53.633
value
P-value 0

Test: Ho: difference in coefficients not systematic

chi2(7) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 53.63
Prob>chi2 = 0.0000
 Stepwise regression in fixed effect model
(1) (2) (3) (4) (5) (6) (7)
Variables
Leverage Leverage Leverage Leverage Leverage Leverage Leverage

150.531*** 143.688*** 49.544** 48.714** 46.337** 45.957** 45.904**

ROA (20.479)
(11.94) (13.669) (22.163) (21.381) (20.763) (20.414)

-12.488 -128.785*** -126.347*** -127.135*** -126.941*** -127.002***

Investment
(25.6)
(11.073) (25.837) (26.454) (25.383) (25.392)

77.749*** 76.258*** 76.468*** 76.38*** 76.421***

Cash_holding
(15.583)
(15.749) (16.133) (15.453) (15.449)

2.831 3.238 3.56 3.56

NWC
(2.78)
(2.697) (2.703) (2.776)

-6.138* -7.433** -7.436**

l_firm_size
(3.363)
(3.288) (3.381)

.436 .437
Firm_Age
(.483)
(.485)

.051
Sales_growth
(1.205)

-12.027*** -10.15*** -.268 -.427 46.611* 42.606 42.607

_cons
(27.91)
(1.523) (2.372) (2.583) (2.604) (25.837) (27.879)

Observations 598 598 598 598 598 598 598

R-squared .931 .935 .977 .977 .978 .978 .978