Scribd
Upload
3 views2 pages

Fixed-Effects Regression on Bank Risk

The fixed-effects regression analysis indicates that high regulatory pressure significantly increases the ratio of non-performing loans to total loans, while the effects of low regulatory pressure and lagged loan-to-deposit ratio are insignificant. Additionally, increases in bank size, return on assets, and return on investment show varying impacts on bank risk, with the latter two reducing non-performing loans. Overall, the model explains 17.96% of the variation in non-performing loans, with statistically significant results for the included variables.

Uploaded by

samuel mensah
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
3 views2 pages

Fixed-Effects Regression on Bank Risk

The fixed-effects regression analysis indicates that high regulatory pressure significantly increases the ratio of non-performing loans to total loans, while the effects of low regulatory pressure and lagged loan-to-deposit ratio are insignificant. Additionally, increases in bank size, return on assets, and return on investment show varying impacts on bank risk, with the latter two reducing non-performing loans. Overall, the model explains 17.96% of the variation in non-performing loans, with statistically significant results for the included variables.

Uploaded by

samuel mensah
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

b.

Assume the individual effects, 𝛼𝑖 is fixed, estimate the model using fixed-effects (FE)
regression. Interpret results.
. xtreg npl_gl hrp lrp [Link] [Link] [Link] [Link], fe

Fixed-effects (within) regression Number of obs = 210


Group variable: bank_id Number of groups = 30

R-sq: Obs per group:


within = 0.3864 min = 7
between = 0.0788 avg = 7.0
overall = 0.1796 max = 7

F(6,174) = 18.26
corr(u_i, Xb) = -0.4461 Prob > F = 0.0000

npl_gl Coef. Std. Err. t P>|t| [95% Conf. Interval]

hrp .0095876 .0029461 3.25 0.001 .0037729 .0154022


lrp .0030224 .0060397 0.50 0.617 -.0088981 .014943

ld
L1. -.0057094 .0081475 -0.70 0.484 -.0217901 .0103713

logsize
L1. .0691095 .0204163 3.39 0.001 .028814 .1094049

roa
L1. -1.536456 .8009201 -1.92 0.057 -3.117226 .0443128

roe
L1. -.2236266 .0950975 -2.35 0.020 -.4113197 -.0359335

_cons -.5598911 .2269054 -2.47 0.015 -1.007732 -.1120499

sigma_u .08917877
sigma_e .07099059
rho .61211032 (fraction of variance due to u_i)

F test that all u_i=0: F(29, 174) = 3.20 Prob > F = 0.0000

Interpretation of results

The coefficient of high regulatory pressure is positive and statistically significant at 1% level,
implying that an additional unit increase in high regulatory pressure result in increase in the ratio
of non-performing loan to total loans by on average, 0.0096, ceteris paribus. Thus, high
regulatory pressure increases bank risk for taking portfolio investment.

The coefficients for low regulatory pressure and first period lag of loan to deposit ratio are
negative and statistically insignificant.
The coefficient of first period lag of the log of bank size is positive and statistically significant at
1% level. This means that a 1% increase in the first period lag of bank size result in increase in
the ratio of non-performing loan to total loans by on average, 0.00069 units, ceteris paribus. The
coefficient is economically insignificant and showing that the first period lag of bank size has
positive but weak effect on bank risk for taking portfolio investment.

The coefficient of first period lag of return on assets is negative and statistically significant at
10% level. This means that a 1 unit increase in the first period lag of return on assets result in
decrease in ratio of non-performing loan to total loans by on average, 1.54, holding other factors
constant. Thus, increase in return on assets reduces bank risk for taking portfolio investment.

The coefficient of first period lag of return on investment is negative and statistically significant
at 5% level. This means that a 1 unit increase in the first period lag of return-on-investment result
in decrease in the ratio of non-performing loans to total loans by on average, 0.2236, holding
other factors constant. Thus, an increase in return on investment reduces bank risk for taking
portfolio investment.

The results also show that 17.96% of the variation in the ratio of non-performing loan to total
loans is explained by the regression. The F-statistic is 18.26 and the associated p-value is 0.0000,
implying that the coefficients in the estimated model are jointly significant.

You might also like