Indian Bank NPAs Dataset (2004–2024)
Description
This dataset supports the analysis presented in the paper "Post-COVID-19 Trends in Non-Performing Assets: Insights from the Indian Banking Sector." It includes unbalanced panel data from 2004 to 2024 covering four bank categories (public sector, private sector, foreign, and scheduled commercial banks) in India. Key variables include Gross and Net NPAs, their ratios to total assets and advances, annual growth rates, and macroeconomic controls (GDP growth, inflation, repo rate). Source: Reserve Bank of India and World Bank.
Files
Steps to reproduce
Download the dataset (Annex_A_Final.xlsx) and open it in Excel, R, Stata, Python, or any preferred statistical software that supports spreadsheet formats. Review the data structure: Panel format with rows representing years (2004–2024) and columns for key variables by bank group. Bank groups include: Public Sector Banks, Private Sector Banks, Foreign Banks, All Scheduled Banks. Key variables include: Gross NPAs (in million INR) Net NPAs (in million INR) Ratio of Gross/Net NPAs to Total Assets (RGNTA) Ratio of Gross/Net NPAs to Advances (RGNPA) Annual Growth Rate (AGR) Macroeconomic controls: GDP Growth, Inflation, Repo Rate Replication of Analyses: Descriptive statistics can be reproduced using summary functions in Excel, R, or Stata. Trend analysis and regressions (OLS and System GMM): Run time series regressions with NPAs as the dependent variable and Year as the independent variable. For dynamic panel estimation, use System GMM with one-period lags of NPAs and macroeconomic controls. Software options: Stata: xtabond2 R: plm or pgmm package Python: linearmodels module with PanelOLS or GMM framework Diagnostics: Validate model assumptions using residual plots, Q–Q plots, and test for autocorrelation and instrument validity (e.g., Arellano-Bond test, Hansen J-test).
Institutions
- Universita Ca' Foscari