Skip to main content

Table 6 Regression scores

From: Corruption and its diverse effect on credit risk: global evidence

Variables

Panel OLS

GMM

Coefficient

t-statistics

Coefficient

t-statistics

Corruption

0.103

2.778 (***)

0.016

0.462

NPLt−1

0.843

74.461 (***)

Capitalization

0.497

31.146 (***)

0.132

12.114 (***)

Credit Disclosure Index

0.079

2.361 (**)

− 0.078

− 3.787 (***)

GDP Growth

− 0.068

− 4.726 (***)

− 0.043

− 4.827 (***)

Inflation

− 0.004

− 0.880

− 0.001

− 0.171

Public Debt

− 0.001

− 0.216

− 0.001

− 0.370

Remittance

0.015

1.213

− 0.003

− 0.004

Trade Openness

0.004

2.511 (**)

0.001

0.628

Unemployment

0.061

4.608 (***)

0.012

1.523

Constant

0.095

0.340

0.126

0.462

Adjusted r-square

0.289

0.746

F-value

145.71 (***)

3013^ (***)

Observations

3200

2844

Hausamn

58.05 (***)

  1. We have performed both fixed and random effect regression based on the model: \({\text{NPL}}_{it} = \alpha_{i} + \beta_{1} {\text{Corruption}}_{it} + \mathop \sum \nolimits_{i = 1}^{i} \beta_{2} {\text{Controls}}_{it} + \varepsilon_{it}\). We have also tried to capture the lag impact of NPL using the GMM model: \(Y_{it} = \alpha + \beta Y_{i,t - 1} + \gamma {\text{ Corruption}}_{it} + \mathop \sum \nolimits_{k = 1}^{k} \delta_{k} {\text{Corruption}}_{t}^{k} + \varepsilon_{it}\). Here, NPL is the dependent variable and is measured by the ratio of non-performing loans to total gross loans. A detailed description of the measurement variable and control variables are provided in Table 4. Hausman test score indicates that the fixed effect model is appropriate for the study. Therefore, we report fixed effect regression scores in Table 6.
  2. ^ represent j-statistic score with the related p value in the parenthesis. Asterisk ** and *** represent significance level at 5 and 1% respectively.