Giới thiệu tài liệu
Research on bank performance using statistical analysis, focusing on fixed effects and random effects models, Hausman test, Wooldridge test for autocorrelation, and FGLS model to examine relationships between various variables such as lnPF and lnIITA. Results highlight the minor relationship between the economic theories and lnIITA, useful for prediction or explaining bank operations.
Đối tượng sử dụng
Researchers, financial analysts, and policymakers interested in understanding bank performance using economic theories
Nội dung tóm tắt
1. **Model Fixed Effect**: The results show that the model can assess bank-specific fixed effects and evaluate variables related to lnIITA. It was found that lnPF, lnPL, and lnLTA have a supportive relationship with lnIITA, while lnDTF and lnLDR exhibit conflicting relationships.
2. **Model Random Effect**: The results of this model are similar to the Fixed Effect Model, as it shows consistent supportive and conflicting relationships between lnPF, lnPL, lnLTA, lnDTF, and lnLDR with lnIITA.
3. **Hausman Test**: The Hausman test compares the results of Fixed Effect and Random Effect models to determine any systematic differences between them. The results indicate no significant difference (p-value = 0.4934), suggesting that both models provide similar information.
4. **Wald Test for Groupwise Heteroskedasticity**: This test checks for the correlation between residuals and their variance in different groups (banks). The results show that the variance is not corrected in the Fixed Effect Model (p-value = 0.0000), indicating that the model cannot account for this factor.
5. **Wooldridge Test for Autocorrelation**: This test examines autocorrelation within the results. The results show some level of autocorrelation (p-value = 0.0224) in the Fixed Effect Model, suggesting that the model may not be properly adjusted for autocorrelation.
6. **FGLS Model**: As a combined approach of Generalized Least Squares, this model can evaluate variables while accounting for both the issues found in the Fixed Effect Model (heteroskedasticity and autocorrelation). The results show that the model exhibits similar characteristics as lnPF, lnPL, and lnLTA with lnIITA, but displays conflicting relationships for lnDTF and lnLDR.
Overall, the research findings demonstrate a minor relationship between economic theories and lnIITA, which can be useful for prediction or explanation of bank operations.
