Factors affecting the ability of small and medium enterprises in Ho Chi Minh City to access bank credit - 14


APPENDIX 3: STATISTICAL RESULTS


Descriptive statistics



. tab covaykhong, sum(namhoatdong)



can't

Summary Mean

of namhoatdong Std. Dev.


Freq.

0

7.1571429

2.8059756

70

1

10.905109

5.0597652

137

Total

9.6376812

4.7642612

207

Maybe you are interested!

Factors affecting the ability of small and medium enterprises in Ho Chi Minh City to access bank credit - 14


. tab covaykhong, sum(experience)



can't

Summary Mean

of experience Std. Dev.


Freq.

0

6.9

3.4099173

70

1

6.5766423

3.6776804

137

Total

6.6859903

3.5843439

207


. tab covaykhong, sum(thkhoan)



can't

Summary of thkhoan Mean Std. Dev.


Freq.

0

.23716215

.19513244

70

1

.03455802

.02515387

137

Total

.1030715

.14967687

207


. horizontal tab


industry


Freq.


Percent


Cum.

0

94

45.41

45.41

1

113

54.59

100.00

Total

207

100.00



. target tab


purpose

Freq.

Percent

Cum.

0

55

26.57

26.57

1

152

73.43

100.00

Total

207

100.00



sum total


Variable


Obs


Mean


Std. Dev.


Min


Max

quymo

207

4.823459

.7297772

3.100371

6.552831


. total revenue


Variable


Obs


Mean


Std. Dev.


Min


Max

revenue

207

4.759144

.5336258

3.121888

5.598375


. sum of balance


Variable


Obs


Mean


Std. Dev.


Min


Max

thought

207

.1030715

.1496769

.0012582

.8524416


Probit model results



. dprobit covayno industry activity experience scale revenue target


Iteration 0: log likelihood = -132.44074

Iteration 1: log likelihood = -68.157496

Iteration 2: log likelihood = -49.128821

Iteration 3: log likelihood = -37.599826

Iteration 4: log likelihood = -34.508679

Iteration 5: log likelihood = -34.046829

Iteration 6: log likelihood = -34.025881

Iteration 7: log likelihood = -34.025821

Probit regression, reporting marginal effects Number of obs = 207

LR chi2(7) = 196.83

Prob > chi2 = 0.0000

Log likelihood = -34.025821 Pseudo R2 = 0.7431


covayk~g

dF/dx

Std. Err.

z

P>|z|

x-bar

[ 95%

CI ]

namhoa~g

.0665679

.0234375

2.81

0.005

9.63768

.020631

.112505

industry

.2676872

.1378453

1.87

0.062

.545894

-.002485

.537859

awesome

-.0741797

.0304896

-2.42

0.015

6.68599

-.133938

-.014421

quymo

.259299

.1157564

2.24

0.025

4.82346

.032421

.486177

revenue

.3624341

.1850911

1.96

0.050

4.75914

-.000338

.725206

thought

-9.643855

1.820189

-5.56

0.000

.103071

-13.2114

-6.07635

purpose

.2418068

.1372135

1.69

0.091

.7343

-.027127

.51074

obs. P

.6618357







pred. P

.5222795

(at x-bar)

( ) dF/dx is for discrete change of dummy variable from 0 to 1

z and P>|z| corresponds to the test of the underlying coefficient being 0



. fit


Probit model for covaykhong, goodness-of-fit test


number of observations =

207

number of covariate patterns =

207

Pearson chi2(199) =

126.45

Prob > chi2 =

1.0000


. lstat


Probit model for covaykhong


True

Classified

D

~D

Total

+

-

136

1

10

60

146

61

Total

137

70

207


Classified + if predicted Pr(D) >= .5 True D defined as covaykhong != 0

Sensitivity

Pr( +| D)

99.27%

Specificity

Pr( -|~D)

85.71%

Positive predictive value

Pr( D| +)

93.15%

Negative predictive value

Pr(~D| -)

98.36%

False + rate for true ~D

Pr( +|~D)

14.29%

False - rate for true D

Pr( -| D)

0.73%

False + rate for classified +

Pr(~D| +)

6.85%

False - rate for classified -

Pr( D| -)

1.64%

Correctly classified


94.69%

Comment


Agree Privacy Policy *