Explore publications.

This paper examines the impact of market share and capital adequacy on the bankruptcy rate of community banks in Tanzania, for the period spanned for 16 years from 2006 to 2021. For instance, recently in Tanzania, seven community banks were closed by the Regulator while others were being merged to increase their going concern and operational efficiency. This study has employed a quantitative approach based on secondary data collected from the Bank of Tanzania as a Regulator and some of them were gathered from community banks. The sample size is 11 community banks, with a total of 176 observations. The panel data analysis has been conducted using statistical software, namely, STATA version 17 with the random effect model being used to generate regression results ready for interpretation. The research findings depicted significant positive correlations between bankruptcy rates as computed using Altman’s Z-score model and the independent variables, namely, market share and capital adequacy ratio. In that regard, the study recommends adequate capital levels and market share to reduce the bankruptcy rates of community banks in Tanzania.

This study aims to enhance credit risk identification, improve loan borrower review efficiency, and increase default prediction accuracy rate using data mining and machine learning techniques. The study also employs deep learning to develop a consumer loan default prediction model that minimizes credit risks and ensures consistent development. The researchers collected data from a survey of 1000 participants, stratified into local and foreign banks, and selected the top 11 banks based on turnover and customer volume. To construct the machine learning model, Keras, a neural network library that runs on TensorFlow, was utilized. The model predicts loan applicant default likelihood. The study's practical implications demonstrate a noteworthy success rate of customer default prediction, which can significantly benefit banks. The model was evaluated on a test set of 250 records and achieved a test set accuracy of 95.2%, correctly predicting the default state of 238 out of 250 respondents.