Savings and Credit Cooperative Organizations (SACCOs) play a pivotal role in promoting financial inclusion, reducing poverty, and supporting social welfare especially in rural and underserved areas. However, 21% of DT-SACCOs do not operate with prudent financing decisions exposing themselves to financial stress and economic shocks. Even among the SACCOs that met compliance requirements, a drop in the capital adequacy ratio from 16.4% in year 2022 to 16.1% in year 2023 signaled alarming financial strain posing a threat to the existing SACCOs. Alarmingly, 35% of DT-SACCOs have ceased operations attributable to improper financing decisions with three delicensed in January 2025, raising significant concerns over their long-term financial health. Thus, the current study aimed to assess the moderating effect of SACCO size on the relationship between financing decision practices and the financial sustainability of Deposit-Taking Savings and Credit Cooperative Organizations (DT-SACCOs) in Kenya. Anchored on the pecking order theory, the research adopted a positivist paradigm and a cross-sectional survey design. A total of 176 finance managers representing 176 licensed DT-SACCOs constituted the study population. Data were collected by structured questionnaires with a 98% response rate as a sample of 122 respondents was selected by Yamane’s formula. Results from a binary logistic regression indicated that introducing the moderator led to a slight increase in the Nagelkerke R², while the inclusion of the interaction terms further strengthened the relationship between predictor variables and financial sustainability. The findings confirmed that SACCO size had a statistically significant moderating effect on this relationship. This study recommends integrating scenario-based stress testing into financing decisions to assess their long-term impact on different funding structures, so as to facilitate their confrontation of different economic conditions.
This study investigates the influence of lunar phases, behavioral indicators, and technical analysis on cryptocurrency returns, with a focus on Bitcoin (BTC) and Ethereum (ETH). Positioned within the context of behavioral finance and astrofinance, the research aims to explore the role of non-traditional factors in shaping market dynamics. The study employs a combination of event study methodology, regression analysis, and machine learning techniques, particularly XGBoost classification, to examine the impact of lunar phases (such as full and new moons), sentiment measures (derived from Google Trends (GT), the Fear & Greed Index (FGI), and Natural Language Processing (NLP)-based sentiment scores), and technical indicators (such as Relative Strength Index (RSI), Moving Average Convergence Divergence (MACD), and Exponential Moving Average (EMA)) on cryptocurrency price movements. Although no statistically significant direct correlation was found between lunar phases and daily returns, it was observed that these phases are associated with changes in investor sentiment and trading volume. Moreover, behavioral variables, including NLP-derived sentiment scores and the FGI, exhibited interaction effects with market returns, particularly during periods of heightened market sentiment. The XGBoost model demonstrated a predictive accuracy of up to 63% for Ethereum, indicating its effectiveness in capturing complex, non-linear relationships within the data. These findings suggest that integrating astrofinancial timing with behavioral signals can improve the predictive accuracy of short-term cryptocurrency market trends, especially during periods of increased sentiment. This research highlights the potential of incorporating alternative data sources, such as lunar events and sentiment indices, into cryptocurrency trading models. The results also emphasize the effectiveness of machine learning algorithms, like XGBoost, in leveraging such non-traditional indicators to optimize investment strategies.