Predicting the financial performance of microfinance institutions with machine learning techniques

The study was undertaken to investigate how corporate governance practices can influence the financial performance of microfinance institutions with a case study of microfinance inking ltd. This study specifically analyzed the effect of board size, gender diversity and CEO duality in the governance structures on performance of microfinance institutions. The descriptive method was used to collect data and the target population was forty-six, where 35 were the staff of INKINGI microfinance ltd 11 were members of the board. In data collection at the field, questionnaire and face-to-face discussion were used. To confirm the reliability, the research conducted a pre-test exercise before the actual data collection process. The validity was tested through experts in the domain of financial statements analysis and managerial decision making by asking the supervisor and other lecturers from Mount Kenya University. The researcher used the SPSS as tool of analysis, which helped to summarize the primary data into quantitative data and the researcher given the proper interpretation of the results basing on research objectives and questions. The analysis showed that board size, board independent and CEO duality were positively correlated with financial performance of Rwandan microfinance institutions (r= 0.810, p< 0.00 r= 00.706, p< 0.04 r= 0.623, p< 0.003 and r= 0.431, p< 0.00) respectively. Furthermore, the research showed that the board independence, gender diversity as well as board size have a considerable effect on financial contribution. According to the perception of the respondents, the results showed that corporate governance increases the profitability of the company, customer loyalty as well as extension of product line and services in the company. The study concluded that board size, board independent and CEO duality are important aspects to consider for better governance of microfinance. The study recommended that INKINGI microfinance ltd should improve their gender diversity for better corporate governance that provides diverse viewpoints, values and new ideas to the boards and provokes lively boardroom discussions. The management should appoint more women directors as they possess managerial skills like public relations, human resource and communication skills than operating and marketing skills. Keywords: Corporate governance practices, financial performance, microfinance institutions, Inkingi Ltd, Rwanda

The subject of this study is the relationship between financial transparency and the financial performance of microfinance institutions (MFIs) in Ethiopia. The research focuses on four core dimensions of transparency: external auditing, public disclosure, compliance with accounting standards, and timeliness of financial reporting, examined through the theoretical frameworks of agency theory and stakeholder theory. The purpose of the study is to determine how these transparency dimensions influence institutional profitability, measured by return on assets (ROA), in a donor-dependent and regulation-intensive environment. The relevance of the research arises from persistent governance challenges, inconsistent reporting practices, and the strategic importance of MFIs in promoting financial inclusion and socio-economic development in emerging markets. The scientific novelty lies in the development of a hybrid conceptual framework that integrates international theoretical models with Ethiopia-specific institutional conditions and in the application of a mixed-methods approach that combines econometric analysis with qualitative insights from industry executives. The methodology includes fixed-effects regression analysis of a balanced panel of 13 licensed MFIs for the period 2015–2024, supported by thematic analysis of semi-structured interviews. The results demonstrate that external auditing, public disclosure, and compliance with accounting standards significantly enhance ROA, while delays in financial reporting reduce profitability. Qualitative findings confirm that transparency improves institutional credibility, donor trust, and operational efficiency, but its implementation is constrained by capacity limitations, regulatory inconsistencies, and divergent stakeholder expectations. The authors conclude that transparency should be viewed not only as a compliance obligation but also as a strategic driver of resilience, competitiveness, and sustainable growth in Ethiopia’s microfinance sector.

This thesis aims to enhance the understanding of microfinance institutions (MFIs) by empirically analyzing their performance in a cross-country setting. It focuses on three research projects, in which the first two analyze the relative performance of these institutions in a dual financial system where microfinance institutions co-exist with commercial banks. The third project analyzes the impact of different CEO attributes on the financial performance of MFIs. The findings of the thesis provide important insights and offer various implications to policymakers, regulators, and MFIs. Policymakers must begin to take steps to incorporate microfinancing into a country's mainstream financial system. The operations and branches of MFIs must be opened in places where they have a market niche or where commercial banks cannot support low-income borrowers. Additionally, MFIs could become more active with start-ups and small-scale enterprises, an operation that has the potential to boost economic growth. Furthermore, policymakers and regulators must introduce new technical ways such as mobile banking and virtual branch networks to minimize the operational cost of microfinance banks and enhance their penetration in remote areas. Likewise, policymakers and regulators should consider adopting systematic risk management techniques such as credit scoring, computerized databases of borrowers' credit histories, loan delinquency rates, and default records to increase the asset quality of these institutions. Finally, MFIs must consider the demographic characteristics when identifying and selecting suitable candidates for the position of CEO. Research findings indicate that MFIs that hire female CEOs, CEOs with business degrees, CEOs with domain-specific experience, and CEOs who are also the founders of MFIs are more likely to increase the financial performance of these institutions.

Sound liquidity management is integral for any financial institution’s stability and profitability, since deteriorating liquidity management is the most frequent cause of poor financial performance. As with any financial institution, the biggest risk in microfinance sector is lending money and not getting it back leading to liquidity problems as most of them have no access to lender of the last resort which is the Central Bank of Kenya. The study sought to investigate the effect of liquidity management on financial performance of microfinance institutions in Kenya. The target population of the study was all the twenty-six microfinance in Kenya that are members of Association of Microfinance Institutions and were licensed by the Central Bank of Kenya as at 2017. A census of all the twenty-six 26 Microfinance Institutions in Kenya was conducted for five years from 2012 to 2016. Secondary data on the study variables was gathered from the audited financial statements of the Microfinance Institutions. The study employed random effect model on a 5-year panel data from 2012 to 2016 on all the 26 Microfinance Institutions in Kenya. The study found a positive relationship between capital adequacy and financial performance and a negative relationship between asset quality, maturity gap and financial performance. The study would help Microfinance Institutions as they would use the research findings to develop liquidity management strategies to enable Microfinance Institutions improve on their financial performance.

This study aims to analyse the financial performance of microfinance institutions (MFIs) in Afghanistan for an extremely unstable period, for the financial years of 2014–2015 to 2023–2024, to address significant political instability and the COVID-19 pandemic. Using panel data from the top larger MFIs in Afghanistan, the study looks at the effects of internal institutional characteristics on the return on assets (ROA), which is an indicator of financial sustainability. Panel regression techniques are employed to analyse the relationships between these factors and institutional performance. The findings indicate that branch expansion enhances ROA; therefore, increasing branches led to greater profitability, even amid political instability and uncertain conditions. While the gross loan portfolio had a negatively significant effect on financial performance, this finding indicates that quality and assessing the associated risk of the loan portfolio may be challenging. Total income and the number of active borrowers were less significant or impactful variables. Overall, the findings indicate that MFIs encountered financial problem and even some of them led to stop their services, moreover, highlighting the need for risk-adapted progress and efficient portfolio control to maintain the financial viability of MFIs working especially in fragile or crisis-affected economies.

We investigate board and management gender diversity issues in the microfinance setting with data (2010–2014) drawn from 494 microfinance institutions across 76 countries. We find that board gender diversity positively predicts management gender diversity. On the effects of board and management gender diversity on the financial performance of microfinance institutions (MFIs), we find that whereas board gender diversity is negatively and significantly related to MFI financial performance, management gender diversity is negatively but insignificantly related to MFI financial performance. We show that 50% or higher diversity in either board or management is the threshold at which gender diversity is productive to MFIs. However, danger exists that an MFI that combines 50% or higher female representation on its board with 50% or higher female representation on its management team is likely to experience a tumble in its financial performance. The overall effect of these outcomes is that the push for more female representation on boards and management teams of MFIs should be done with a lot of tact and circumspection.

This paper assesses the impact of information and communication technologies (ICTs) on the performance of microfinance institutions (MFIs) in Niger, West Africa. MFIs play a pivotal role in improving financial inclusion in Niger because the majority of the country’s poor live in rural areas, with only limited and costly access to formal financial services. Using an unbalanced panel of 23 MFIs spanning 2005–2013, single-step generalized moments method (GMM) estimations are run to appraise whether ICT investments improve the financial and the social performance of MFIs. The results show a positive relationship between investments in ICTs and MFIs’ financial performance. Investing more in technologies enables managers to reduce the frequency of operational errors, increase the speed of task execution, decrease operating costs, and increase the likelihood of higher financial profits. The findings also reveal a positive effect of institutional affiliation on the financial performance of MFIs. Namely, MFIs affiliated with a network and investing in ICTs tend to perform better. The impact of ICT investments on the social performance of MFIs is rather weak. From a policy perspective, developing ICT infrastructure can yield substantial performance dividends and should remain a top developmental priority in Niger.

This study uses machine learning (ML) techniques to predict maximum flank wear on cutting tools during the turning of Ti6Al4V, a titanium alloy known for its challenging machinability. We used three models (a) support vector machines (SVM), (b) random forests (RF), and (c) neural networks (NN) to predict maximum flank wear. The machining parameters are taken as input data sets, which compromise cutting speed, feed, and machining time. All the experiments are done at a constant depth of cut to predict maximum flank wear as an output variable. The forecasted maximum flank wear was validated using experimental data to verify their precision. The SVM model outperformed the other ML models in a wider range of cutting parameters concerning tool flank wear. The SVR model is the most accurate model with an average forecasting error of 3.24%. In forecasting maximum flank wear, the RF and NN models followed, with average forecasting errors of 3.88% and 3.71%, respectively. Additionally, the SVM model's robustness was further validated by its ability to maintain stability at varying cutting speeds and feed rates. The research enhances the economics of the machining process by predicting tool flank wear and fostering sustainability in the manufacturing industry, thereby demonstrating the applicability of ML techniques.

Purpose The purpose of this study was to investigate on managing market risk and financial performance, experience from microfinance institutions (MFIs) in Kenya. Design/methodology/approach This study used positivism philosophy and used explanatory non-experimental research designs. The targeted population was all the 13 registered deposit-taking MFIs in Kenya and a census approach was used. The study used secondary data which was collected and analyzed from microfinance Institutions annual audited financial reports for the period between 2014 to 2018. This study was anchored on two theories, namely, resource-based value theory and extreme value theory. Findings The results indicated that interest rate and financial leverage risk had a positive significant effect on the financial performance of MFIs in Kenya. Foreign exchange risk was found to have a negative significant effect on the financial performance of MFIs. However, inflation rate risk was found to have no significant effect on the financial performance of MFIs. Research limitations/implications This study recommended that the chief executive officers of MFIs should use the mechanism of identifying market risk variables, especially Interest rate, financial leverage and foreign exchange risks to enable them to put the necessary measures to mitigate those risks and enhance the financial performance of MFIs in Kenya. Originality/value This study is unique as it touches the microfinance industry which has a steady fast growth in assisting accessibility of financial services to small and medium enterprises. Most of the previous study concentrated on other industry in the financial sector.

BackgroundOvarian clear cell carcinoma (OCCC) represents a subtype of ovarian epithelial carcinoma (OEC) known for its limited responsiveness to chemotherapy, and the onset of distant metastasis significantly impacts patient prognoses. This study aimed to identify potential risk factors contributing to the occurrence of distant metastasis in OCCC.MethodsUtilizing the Surveillance, Epidemiology, and End Results (SEER) database, we identified patients diagnosed with OCCC between 2004 and 2015. The most influential factors were selected through the application of Gaussian Naive Bayes (GNB) and Adaboost machine learning algorithms, employing a Venn test for further refinement. Subsequently, six machine learning (ML) techniques, namely XGBoost, LightGBM, Random Forest (RF), Adaptive Boosting (Adaboost), Support Vector Machine (SVM), and Multilayer Perceptron (MLP), were employed to construct predictive models for distant metastasis. Shapley Additive Interpretation (SHAP) analysis facilitated a visual interpretation for individual patient. Model validity was assessed using accuracy, sensitivity, specificity, positive predictive value, negative predictive value, F1 score, and the area under the receiver operating characteristic curve (AUC).ResultsIn the realm of predicting distant metastasis, the Random Forest (RF) model outperformed the other five machine learning algorithms. The RF model demonstrated accuracy, sensitivity, specificity, positive predictive value, negative predictive value, F1 score, and AUC (95% CI) values of 0.792 (0.762–0.823), 0.904 (0.835–0.973), 0.759 (0.731–0.787), 0.221 (0.186–0.256), 0.974 (0.967–0.982), 0.353 (0.306–0.399), and 0.834 (0.696–0.967), respectively, surpassing the performance of other models. Additionally, the calibration curve's Brier Score (95%) for the RF model reached the minimum value of 0.06256 (0.05753–0.06759). SHAP analysis provided independent explanations, reaffirming the critical clinical factors associated with the risk of metastasis in OCCC patients.ConclusionsThis study successfully established a precise predictive model for OCCC patient metastasis using machine learning techniques, offering valuable support to clinicians in making informed clinical decisions.

Posttraumatic stress disorder (PTSD) recently becomes one of the most important mental health concerns. However, no previous study has comprehensively reviewed the application of big data and machine learning (ML) techniques in PTSD. We found 873 studies meet the inclusion criteria and a total of 31 of those in a sample of 210,001 were included in quantitative analysis. ML algorithms were able to discriminate PTSD with an overall accuracy of 0.89. Pooled estimates of classification accuracy from multi-dimensional data (0.96) are higher than single data types (0.86 to 0.90). ML techniques can effectively classify PTSD and models using multi-dimensional data perform better than those using single data types. While selecting optimal combinations of data types and ML algorithms to be clinically applied at the individual level still remains a big challenge, these findings provide insights into the classification, identification, diagnosis and treatment of PTSD.

Assessing treatment switch among patients with multiple sclerosis: A machine learning approach

With the popularization of machine learning (ML) techniques and the increased chipset’s performance, the application of ML to pedestrian localization systems has received significant attention in the last years. Several survey papers have attempted to provide a state-of-the-art overview, but they usually limit their scope to a particular type of positioning system or technology. In addition, they are written from the point of view of ML techniques and their practice, not from the point of view of the localization system and the specific problems that ML techniques can help to solve. This article is intended to offer a comprehensive state-of-the-art survey of the ML techniques that have been adopted over the last ten years to improve the performance of pedestrian localization systems, addressing the applicability of ML techniques in this domain, along with the main localization strategies. It concludes by indicating the underlying open issues and challenges associated with the existing systems, and possible future directions in which ML techniques could improve the performance of pedestrian localization systems. Among other open issues, most previous authors have focused their attention on position estimation accuracy, which wastes the potential of ML techniques to improve other performance parameters (e.g., response time, computational complexity, robustness, scalability or energy efficiency). This study shows that there is a strong trend towards the application of supervised learning. Consequently, there are many potential research opportunities in the use of other learning types, such as unsupervised and reinforcement learning, to improve the performance of pedestrian localization systems.

ABSTRACTThere is some consensus that depth of outreach and financial performance of Microfinance Institutions (MFI) are positively correlated. A majority of microcredit borrowers are women and since the prevalence of female borrowers is even greater among the very poor, there should be a positive correlation between female borrowers and financial performance. Most of the MFIs target women as preferred borrowers. However, no study to date has investigated the relationship between targeting women and MFI’s sustainability with respect to profitability and yield. Utilizing a panel of 892 MFIs over a period of 10 years, this study shows that increased proportion of women borrowers has a statistically significant positive impact on yield and financial performance of MFIs. Consequently, this article also analyses the implication as to whether female borrowers have better repayment rate than male borrowers.

Landfills are one of the major sources of greenhouse gas emissions worldwide. Over the past few decades, gas collection and energy conversion systems have been increasingly implemented in landfills to enhance sustainability and profitability. Accurate methane (CH4) generation rate estimates are essential for designing and optimizing these systems. However, due to the heterogeneity of waste properties, site-specific conditions, and the complex interrelated thermo-hydro-bio-mechanical processes involved, obtaining precise CH4 generation predictions is challenging. Although laboratory and field monitoring data provide valuable insights, obtaining such data is expensive and is often site-specific and non-generalizable. Process-based numerical models, such as the coupled thermo-hydro-bio-mechanical (CTHBM) model developed at the University of Illinois Chicago, offer reliable estimates but are computationally intensive for optimization tasks. Recent advancements in machine learning (ML) based data-driven modeling show promise in handling complex nonlinear datasets. This study evaluates the effectiveness of several ML models, including random forests (RF), extreme gradient boosting (XGB), multi-layer perceptron (MLP), and linear regression (LR), in predicting landfill CH4 generation rates. The data used to train these ML models is generated using the CTHBM model by varying several waste properties. Results indicate that during the rapid methane generation phase, the MLP model demonstrated excellent performance, while XGB also showed reasonable accuracy. In contrast, RF and LR models did not accurately predict methane generation rates during this phase. This study is a preliminary proof of concept for applying ML techniques in landfill CH4 prediction. Further exploration of advanced techniques, such as physics-informed ML, is recommended for scalable and generalizable predictions.