-
Research Article
A Machine Learning-Based Prediction of Malaria Occurrence in Kenya
Issue:
Volume 13, Issue 4, August 2024
Pages:
65-72
Received:
20 July 2024
Accepted:
9 August 2024
Published:
20 August 2024
Abstract: For many years’ malaria has been a health public concern in Kenya as well as many parts of Africa and other parts of the world. The purpose of this study is to develop and evaluate a supervised machine learning model to predict malaria occurrence (final malaria test results) in Kenya. The study investigated twelve predictor variables on the outcome variable (malaria test results), where five machine learning models namely; k-nearest neighbors, support vector machines, random forest, tree bagging, and boosting, were estimated. During the model evaluation, random forest emerged as the best overall model in the classification and prediction of final malaria test results. The model attained a higher classification accuracy of 97.33%, sensitivity of 71.1%, specificity of 98.4%, balanced accuracy of 84.7% and an area under the curve of 98.3%. From the final model, the presence of plasmodium falciparum emerged most important feature, followed by region, endemic zone and anemic level. The feature with the least importance in predicting final malaria test results was having mosquito nets. In conclusion, employing Machine learning algorithms enhances early detection, optimizing resource allocation for interventions, and ultimately reducing the incidence and impact of malaria in the Kenya. The study recommends allocation of resources and funds to areas with the presence of plasmodium falciparum, region susceptible to malaria, endemic zones and anemic prone areas.
Abstract: For many years’ malaria has been a health public concern in Kenya as well as many parts of Africa and other parts of the world. The purpose of this study is to develop and evaluate a supervised machine learning model to predict malaria occurrence (final malaria test results) in Kenya. The study investigated twelve predictor variables on the outcome...
Show More
-
Research Article
Application of Vector Autoregressive (VAR) Model on the Interaction of Inflation Rates and Public Debt in Kenya from 2011 to 2021
John Maina,
Robert Muriungi,
Harun Gitonga*
Issue:
Volume 13, Issue 4, August 2024
Pages:
73-79
Received:
30 July 2024
Accepted:
10 August 2024
Published:
22 August 2024
Abstract: This study examines the relationship between public debt and inflation rates in Kenya from 2011 to 2021 using the Vector Autoregressive (VAR) model. Despite the models likeAutoregressive Integrated Moving Average (ARIMA), Seasonal Autoregressive Integrated Moving Average (SARIMA), and Generalized Autoregressive Conditional Heteroscedasticity (GARCH) gaining popularity in time series analysis, the Vector Autoregressive model, being multivariate, is relevant in analyzing two or more time series variables simultaneously, benefiting from the bi-directional causality and providing a better outlook into the flow of the dynamic interaction between inflation and public debt. The main objectives are modelling the Vector Autoregressive model and forecasting future trends to provide insights for policymakers. Additionally, the methodological approach comprises descriptive statistics, stationarity tests, normality tests, and the Vector Autoregressive model. Descriptive statistics reveal significant variations, with public debt increasing from 1.35 trillion KES to a peak of 8.2 trillion KES, and inflation rates ranging from 3.2% to 19.72% for the period from 2011 to 2021. The Augmented Dickey-Fuller (ADF) test confirmed that both time series were stationary at their levels. The Vector Autoregressive model, chosen for its ability to analyze dynamic interactions, indicated a significant relationship between the variables, with inflation showing strong self-persistence (coefficient of 0.8731, p < 2 × 10−16), though public debt did not significantly impact inflation in the model (p = 0.5592). The models R-squared values, 95.82% for public debt and 84.74% for inflation, highlight its strong explanatory power. Moreover, findings indicate that while public debt does not directly affect inflation within the model lag structure, inflation exhibits a strong self-persistence. The model R-squared values are 95.82% for public debt and 84.74% for inflation, demonstrating high explanatory power. Recommendations include the implementation of a robust debt management strategy, emphasizing sustainable borrowing and enhancing revenue generation to mitigate inflationary pressures. Further research is recommended to explore the broader macroeconomic impacts of public debt on economic growth and employment in Kenya.
Abstract: This study examines the relationship between public debt and inflation rates in Kenya from 2011 to 2021 using the Vector Autoregressive (VAR) model. Despite the models likeAutoregressive Integrated Moving Average (ARIMA), Seasonal Autoregressive Integrated Moving Average (SARIMA), and Generalized Autoregressive Conditional Heteroscedasticity (GARCH...
Show More
-
Research Article
A Generalized Linear Model of HIV/AIDS Patients in Kenya: A Case Study of Nyeri County Referral Hospital
Sarah Ireri*,
Joseph Esekon,
Margaret Kinyua
Issue:
Volume 13, Issue 4, August 2024
Pages:
80-84
Received:
24 July 2024
Accepted:
13 August 2024
Published:
22 August 2024
Abstract: With millions of new cases and deaths reported every year, HIV/AIDS is a significant worldwide health concern. Creating successful public health policies and interventions requires an understanding of the dynamics of HIV transmission and progression. WHO predicted that by the end of 2022 roughly 39 million individuals worldwide would be living with HIV, out of which 37.5 million are adults, whereas 1.5 million are children. Despite outstanding global gains in HIV/AIDS prevention, treatment and care, Kenya continues to struggle to effectively handle the HIV epidemic, particularly in areas like Nyeri County. Nyeri County Referral Hospital is a critical healthcare institution for HIV/AIDS patients in the region. However, there is still a lack of understanding about the epidemiological characteristics of HIV/AIDS in this particular population. This study’s aim was to use a GLM on HIV/AIDS data in Nyeri County Referral Hospital in Kenya. To determine the significance of model parameters, Likelihood Ratio Test was used whereas significance of regression coefficients was determined using Wald Chi- Square Test. Deviance was utilized to test for the goodness of fit. R software version 4.4.1 was utilized. This project may help health policymakers in developing or refining HIV/AIDS care programs. Findings from the study can help healthcare planners and policymakers allocate resources more efficiently to meet the requirements of HIV/AIDS patients. The fitted model showed that, only ART use was significant (p-value = 2.684562 × 10−13). Because some covariates were not significant, each of them was analyzed separately. Age was a significant predictor (p-value = 0.0001536103). The other variables were not significant. This finding is consistent with previous evidence, which stresses the relevance of ART in lowering viral load, enhancing immunological function, and extending the lives of people living with HIV. To build upon the current findings, future research should explore additional variables that may influence HIV status, for example cultural beliefs, and access to healthcare services. Again, future studies may involve the use of survival analysis through GLM in analyzing similar data.
Abstract: With millions of new cases and deaths reported every year, HIV/AIDS is a significant worldwide health concern. Creating successful public health policies and interventions requires an understanding of the dynamics of HIV transmission and progression. WHO predicted that by the end of 2022 roughly 39 million individuals worldwide would be living with...
Show More
-
Research Article
Modelling Petrol Prices in Kenya from 2014 to 2023 Using Sarimax Model: A Case Study of Nairobi County
Fidelis Nyamai,
Joseph Esekon,
Edwine Atitwa*
Issue:
Volume 13, Issue 4, August 2024
Pages:
85-91
Received:
1 August 2024
Accepted:
16 August 2024
Published:
26 August 2024
Abstract: The requirement for petrol price information is crucial for majority of enterprises. This is because fluctuations in petrol prices impact inflation hence affecting daily lives of citizens. In analyzing the prices of petrol, researchers have employed several models but encountered various limitations. These limitations include; the Error Correction Model can examine only one co-integrating association. The Vector Autoregression (VAR) model does not account for the structural changes in the data. Additionally, the AutoRegressive Integrated Moving Average (ARIMA) model does not take into consideration the seasonal component in the data. The Generalized Autoregressive Conditional Heteroskedasticity (GARCH) model assumes that over time the volatility is constant. Moreover, the Seasonal Autoregressive Integrated Moving Average (SARIMA) model does not integrate the external factors. Hence in this study Seasonal Autoregressive Integrated Moving Average with Exogenous Variables (SARIMAX) model was employed since it captures seasonality in data and incorporates the exogenous variables. The research’s aim was to model prices of petrol in Kenya for the period between 2014 to 2023 with exchange rates as an external factor. Secondary data was obtained from Energy and Petroleum Regulatory Authority (EPRA), Kenya National Bureau of Statistics (KNBS) and Central Bank of Kenya (CBK) websites. R software was used to analyze the data. By the use of historical data of petrol prices and exchange rates, the study sought to fit the best Seasonal Autoregressive Integrated Moving Average with Exogenous Variables (SARIMAX) model, validate the model and predict the petrol prices. The petrol price data was found to be non-stationary using Augmented Dickey Fuller test (ADF). Regular differencing was conducted to make the data stationary. Seasonal differencing due to seasonality component available in the data was also performed. Best SARIMAX model was chosen from various SARIMAX models according to Box-Jenkins methodology which uses least Akaike Information Criterion (AIC) value. SARIMAX (0,1,1)(2,1,2)12 model was selected since it had least Akaike Information Criterion (AIC) value of 656.3733 and the model validated using the hold out technique. The forecasts errors from the training set were; Mean Squared Error (MSE)=10.4970, Root Mean Square Error (RMSE)=3.239911, Mean Absolute Percentage Error (MAPE)=2.309268% while those from the testing set were; Mean Squared Error (MSE)=3271.1012, Root Mean Square Error (RMSE)=57.193542, Mean Absolute Percentage Error (MAPE)=26.695390%. There was less error in the training set than in the testing set as it was expected hence the model suited the data well and could be used for future predictions. The model was then used for five year forecast into the future. This study’s findings will offer sound suggestions to policymakers, businesses and consumers. This study recommends a model to be fitted using other factors affecting petrol prices and fitting Fourier terms, Behavioral Assessment Tools (BATS) and Trigonometric Box-Cox ARMA Trend Seasonal (TBATS) models.
Abstract: The requirement for petrol price information is crucial for majority of enterprises. This is because fluctuations in petrol prices impact inflation hence affecting daily lives of citizens. In analyzing the prices of petrol, researchers have employed several models but encountered various limitations. These limitations include; the Error Correction ...
Show More
,
Victor Wandera Lumumba
