SU+ Digital Repository

SU+ is an online repository for the preservation and promotion of assorted digital content at Strathmore University

Off-Campus Access to restriced resources (including the ExamsBank) now requires registration using an @strathmore.edu email address

Authentication is NOT required for On-Campus Access to content

Photo by @inspiredimages

Communities in DSpace

Select a community to browse its collections.

Now showing 1 - 5 of 7

Recent Submissions

  • Thumbnail Image
    Item type:Item,
    Forest acoustics - Temporal Frequency Convolution Neural Network model for detecting illegal logging activities in forest
    (Strathmore University, 2024) Simiyu, Daniel
    Illegal logging has detrimental effects on the environment, including increased deforestation, which hampers carbon sequestration and leads to the heightened carbon emissions. Consequently, there is an imperative to determine and prevent illegal logging to safeguard the environment. Detecting logging activities involves discerning complex environmental sounds. Environmental sounds such as forest acoustics are very complex given that they often have background noises such as wind making it difficult to classify them. This study concentrates on employing deep learning techniques for classifying forest acoustics, tackling the complexities inherent in environmental sounds. It specifically delves into temporal and frequency patterns of forest sounds, isolating sounds indicative of logging activities such as chainsaw sounds. The Temporal Frequency Convolutional Neural Network (TFCNN) is deployed to accurately recognize logging related acoustics while dealing with background noise during classification and data processing. The TFCNN model consists of two main branches, the harmonic spectrogram branch, and the percussive spectrogram branch. The outputs of both harmonic and percussive branches are concatenated to combine the information learned from both. A convolutional layer with 256 filters and a kernel size of (1,1) is added after concatenation. The output is then flattened to prepare for fully connected layers. The model was trained using the FSC 22 dataset, which is tailored for forest – related analyses, making it an ideal choice. To tackle background noise during classification, attention mechanisms and feature representation techniques were employed. The model recorded an accuracy of 87%. Generally, the model was able to identify illegal logging sounds while dealing with background noise during classification. An android mobile application was developed to serve as the user interface for capturing audio input. Overall, the model properly identifies illegal logging sounds amidst varying levels of background noise. Keywords: Attention Mechanism, Feature Representation, Deep Leaning, Illegal Logging.
  • Thumbnail Image
    Item type:Item,
    The Moderating role of bank size on sovereign debt exposure and financial stability of commercial banks in Kenya
    (Strathmore University, 2026) Njuguna, Judy Wambui
    The growing exposure of commercial banks to sovereign debt in developing economies has raised critical concerns regarding financial stability, particularly within the context of Kenya’s evolving fiscal landscape. This study addressed this gap by examining the moderating role of bank size on the relationship between sovereign debt exposure and the stability of commercial banks in Kenya. The general objective was to determine how bank size conditions the effects of sovereign debt exposure on bank stability, while the specific objectives focused on sovereign debt concentration, crowding-out of private credit, and sovereign credit ratings. The study was anchored on the sovereign– bank nexus (doom loop) theory and moral hazard theory, which explain the interconnected risks between government debt and banking sector stability. A positivist research philosophy and an explanatory research design were adopted to facilitate objective hypothesis testing. The target population comprised 39 licensed commercial banks in Kenya, from which an unbalanced panel dataset of 507 observations covering the period 2010–2024 was constructed using secondary data obtained from audited financial statements and Central Bank of Kenya reports. The study employed panel data econometric techniques, specifically the Random Effects Generalized Least Squares (GLS) model. Bank stability was measured using the Z-score, while sovereign debt exposure was proxied through sovereign debt concentration, private sector loans to total assets, and sovereign credit ratings, with bank size introduced as a moderating variable. The findings revealed that sovereign debt concentration significantly reduces bank stability, while sovereign credit ratings and profitability enhance stability. Bank size exhibited a significant moderating effect, indicating that the impact of sovereign exposure varies across large, medium and small banks. The results also showed that asset quality deterioration weakens stability, while crowding-out effects were not statistically significant in the direct model. The study concluded that excessive sovereign debt exposure poses systemic risks to the banking sector, and that bank size plays a critical role in shaping resilience. Policy recommendations include strengthening macroprudential regulations to limit excessive sovereign exposure, enhancing capital requirements, and adopting differentiated regulatory frameworks that account for bank size heterogeneity. Key words: Sovereign debt exposure, sovereign debt concentration, Sovereign bank nexus, bank stability, bank size,
  • Thumbnail Image
    Item type:Item,
    Effect of deposit insurance on stability of commercial banks in Kenya: moderating role of bank size
    (Strathmore University, 2026) Fardosa, Hassan
    Bank stability is fundamental to financial system resilience, depositor confidence, and sustainable economic growth, particularly in emerging banking systems. This study examined the effect of deposit insurance on the stability of commercial banks in Kenya and the moderated role of bank size, focusing on deposit insurance coverage, depositor protection, and deposit insurance funding. The main objectives of the research were to determine the effect of deposit insurance coverage on the stability of commercial banks in Kenya, to establish the effect of depositor protection on the stability of commercial banks in Kenya, to assess the effect of deposit insurance funding on the stability of commercial banks in Kenya, and to determine the moderating effect of bank size between deposit insurance and bank stability. Anchored in moral hazard and bank-run theories, the study adopted a positivist research philosophy and a quantitative, descriptive research design. Using an unbalanced panel dataset of 38 licensed commercial banks from 2015 to 2024, secondary data were sourced from banks’ audited financial statements, the Central Bank of Kenya, and the Kenya Deposit Insurance Corporation. Data analysis combined descriptive statistics with panel econometric techniques. Fixed Effects and Random Effects models were estimated, with selection guided by the Hausman specification test and supported by robustness analysis using Feasible Generalized Least Squares (FGLS). The empirical findings reveal that deposit insurance coverage and depositor protection do not exert a statistically significant within-bank effect on stability once unobserved heterogeneity is controlled for under the Fixed Effects specification. However, robustness analysis using FGLS indicates that depositor protection is positively and significantly associated with asset quality ratio, suggesting potential market-discipline weakening effects consistent with moral hazard dynamics. Deposit insurance funding emerged as a relevant dimension, exhibiting a weak but positive association with asset quality ratio. These underscores increased risk taking by banks with high deposit insurance funding. Interaction effects reveal that bank size does not moderate the relationship between deposit insurance and stability; however, bank size is a significant direct determinant of stability, reflecting diversification and scale advantages. These findings offer policy-relevant insights for strengthening Kenya’s deposit insurance framework through appropriate design of the DIS and supervisory complementarity to strengthen banking stability. Keywords: Deposit Insurance, Bank stability, Depositor Protection, Moral Hazard, Bank Run
  • Thumbnail Image
    Item type:Item,
    Social factors affecting women groups’ investment behaviour in Nairobi’s Mathare informal settlements
    (Strathmore University, 2026) Ngugi, Regina Wanjiku
    From a world population of slightly above eight billion people as estimated by the United Nations as at November 2022, Informal settlements or slums dominate the urban landscape and were the home to almost One point one two billion people as of the same year. The world population of women in the same year was estimated at 3.95billion. More than 50 per cent of the urban female population this year was estimated to be staying in the informal settlements. Research has shown that the majority population in informal settlements is economically marginalized and experiencing high poverty levels. As such, they depend on informal sources of finance (self-help groups) to meet their financial requirements. The bulk of the studies have focused on savings and lending within women's groups, with minimal consideration of converting these savings into investments, whether individually or as a group. This study aimed to examine the nature of investment exhibited by women in informal settlements and the factors that inform this investment. The research was done in the Mathare informal settlement. It was an explanatory research design study guided by Prospect theory and the Modern Portfolio, using questionnaires and Focus group discussions as means of data collection. Data from the questionnaire was analyzed using a regression model to determine the extent to which each independent variable influenced the dependent variable. In the correlation analysis, all the research variables, such as Education, Spousal Support) and Age, all influenced the investment behaviour of women in informal settlements in Nairobi, Kenya, positively. The results of the regression analyses also indicated that the social factors investigated accounted for the variation among the women in the informal settlements of Mathare in Nairobi. The chi-square test statistics, show that all social factors significantly positively impacted investment behaviour. This research therefore finds that social factors can influence women's behaviour in the informal settlements of Mathare in Nairobi regarding investment. The research further concluded that these factors must be aligned with the particular investment behaviours to enhance investment behaviour. Non-governmental Organisations (NGOs) are encouraged to participate in training and the acquisition of skills needed to handle finances. The research suggests that the government of Kenya's existing funds should be increased to support the improvement of these skills. The research also suggests that women investors should be advised to invest more in their awareness and knowledge of investment trends and opportunities through means such as attending business seminars and engaging investment advisors. Key words: Social factors, Investment behavior, Education level, age, marital status, saving, women groups
  • Thumbnail Image
    Item type:Item,
    Factors influencing the financing choice among creative enterprises in Nairobi County, Kenya: the moderating role of government support
    (Strathmore University, 2026) Kahariri, Kevin Kariuki
    In Nairobi County, Kenya, creative businesses' sustainability and expansion depend critically on their financial decisions. The aim of this investigation was to look at the elements affecting financing choices for creative enterprises. Based on the Life Cycle Theory of Financing and the Pecking Order Theory, the research investigated using government support as a moderating factor how growth possibilities, ownership, asset tangibility, and financial literacy affect financing decisions. Targeting 100 creative businesses across ten creative industry sub-sectors, categorised by UNCTAD, the study used a descriptive correlational research design. Structured questionnaires gathered data; PLS-SEM was used for analysis. This investigation is aimed to provide insightful analysis of the financing practices of creative businesses and guide policy and financial institutions on the creation of initiatives catered to this sector. From the results, the study found out that ownership control determines significantly financing choices among the creative enterprises under study. Growth opportunities is an insignificant determinant of financing choice among the creative enterprises under study. Asset tangibility determines significantly financing choice among the creative enterprises under study. Financial literacy determines significantly financing choice among the creative enterprises under study. The investigation concluded that government support does not have a moderating effect on the determinants of financing choice for creative enterprises. The study recommends that entrepreneurs adopt more formalized and diversified ownership structures. Creative businesses under study should align their financing choices with scalable and sustainable business models. The creative businesses under study should strategically utilize their intangible assets such as intellectual property, brand equity, and digital content as part of their financing strategies. The study further recommended that it is essential for the businesses under study to prioritize financial literacy as a foundational element of entrepreneurial success. Key words: financing choice, growth opportunities, ownership control, tangibility of assets, financial literacy, government support, creative businesses, Nairobi County, Kenya.