Browsing by Author "Ayele, Dawit Getnet."

Now showing 1 - 4 of 4

Longitudinal analysis of the effect of climatic factors on the wood anatomy of two eucalypt clones.
(2010) Ayele, Dawit Getnet.; Zewotir, Temesgen Tenaw.; Ndlovu, Principal.
Eucalypt trees are one of tree species used for the manufacturing of papers in South Africa. The manufacturing of paper consists of cooking the wood with chemicals until obtaining a pulp. The wood is made of different cells. The shape and structure of these cells, called wood anatomical characteristics are important for the quality of paper. In addition, the anatomical characteristics of wood are influenced by environmental factors like climatic factors, soil compositions etc…. In this study we investigated the effects of the climatic factors (temperature, rainfall, solar radiation, relative humidity, and wind speed) on wood anatomical characteristics of two Eucalyptus clones, a GC (Eucalyptus grandis × camuldulensis) and a GU (Eucalyptus grandis × urophylla). Nine trees per clone have been selected. Two sets of data have been collected for this study. The first set of data was eleven anatomical characteristics of the wood formed daily over a period of five years. The second set of data was the daily measurement of temperature, rainfall, solar radiation, relative humidity and wind speed in the experimental area. Wood is made of two kinds of cell, the fibres and the vessels. The fibres are used for the strength and support of the tree and the vessels for the nutrition. Eleven characteristics related to those cells have been measured (diameter, wall thickness, frequency). These characteristics are highly correlated. To reduce the number of response variables, the principal component analysis was used and the first four principal components accounts for about 95% of the total variation. Based on the weights associated with each component the first four principal components were labelled as vessel dimension (VD), fibre dimension (FD), fibre wall (FW) and vessel frequency (VF). The longitudinal linear mixed model with age, season, temperature, rainfall, solar radiation, relative humidity and wind speed as the fixed effects factors and tree as random effect factor was fitted to the data. From time series modelling result, lagged order of climatic variables were identified and these lagged climatic variables were included in the model. To account for the physical characteristic of the trees we included the effect of diameter at breast height, stem radius, daily radial increment, and the suppression or dominance of the tree in the model. It was found that wood anatomical characteristics of the two clones were more affected by climatic variables when the tree was on juvenile stage as compared to mature stage.
Statistical Modeling of Acute HIV Infection from a Cohort of High-risk Individuals in South Africa = Ukufakwa kwamamodeli Ezibalomininingo Ekuthelelekeni Kwesikhashana nge-HIV Eqoqweni labantu Abasengcupheni Enkulu eNingizimu Afrika.
(2022) Yirga, Ashenafi Argaw.; Melesse, Sileshi Fanta.; Mwambi, Henry Godwell.; Ayele, Dawit Getnet.
In this dissertation, longitudinal data modeling approaches to analyze data on CD4 cell counts measured repeatedly in HIV-infected patients enrolled in the Centre for the AIDS Programme of Research in South Africa are investigated. Longitudinal data, or repeated measurement data, is a specific form of multilevel data. In longitudinal studies, repeated observations are made on an individual on one or more outcomes, including covariates information at a baseline and over time. Mixedeffects models have become popular for modeling longitudinal data. This statistical procedure also permits the estimation of variability in hierarchically structured data and examines the impacts of factors at different levels. Since longitudinal studies are often faced with the incompleteness of the data due to partially observed subjects, the mixed-effects model is by its very nature able to deal with unbalanced data of this nature. Therefore, the study adopts the mixed-effects model and identifies whether specific clinical and sociodemographic factors present in the data influenced CD4 count in a cohort of HIV-infected patients. Since it is of great interest for a biomedical analyst or an investigator to correctly model the CD4 cell count or disease biomarkers of a patient in the presence of covariates or factors determining the disease progression over time, the Poisson regression approach, which explain variability in counts, is considered. The Poisson generalized mixed-effects models can be an appropriate choice for repeated count data. However, this model is not realistic because of the restriction that the mean and variance are equal. Therefore, the Poisson mixed-effects model is replaced by the negative binomial mixed-effects model. The later model effectively managed over-dispersion of the longitudinal data. We evaluate and compare the proposed models and their application to model CD4 cell counts of HIV-infected patients recruited in the study data set. The results reveal that the negative binomial mixedeffects model has appropriate properties and outperforms the Poisson mixed-effects model in terms of handling the over-dispersion of the data. Multiple imputation techniques are also used to handle missing values in the dataset to validate parameter estimates in modeling the negative binomial mixed-effects model by assuming a missing at random missingness. To illustrate the full conditional distribution of the repeated outcome, a quantile mixed-effects model is employed. This gives greater inclusive statistical modeling than conventional ordinary mixed models. Quantile regression offers an invaluable tool to discern effects that would be missed by other conventional regression models, which are solely based on modeling conditional mean. The quantile regression model that assumes asymmetric Laplace distribution for the error term was applied to longitudinal CD4 count data. The exact maximum likelihood estimation of the covariate effects and variance-covariance elements in the quantile mixed-effects model was implemented using the Stochastic Approximation Expectation-Maximization algorithm. In the model, multiple random effects are also incorporated to consider the correlation among the observations. Thus, we obtain robust parameter estimates for various conditional distribution positions that communicate an inclusive and more complete picture of the effects. Furthermore, to get more insights into the functional relationship between the response variable and the covariates, the generalized additive mixed-effects models, such as the additive negative binomial mixed-effects model, a versatile model used to better understand and analyze complex nonlinear trajectories in an overdispersed longitudinal data, is applied. Following the additive negative binomial mixed-effects model, an attempt to fit additive quantile mixed-effects model, an efficient and flexible framework for nonparametric as well as parametric longitudinal forms of data analysis focused on features of the outcome beyond its central tendency, was made. The response variable at hand is a CD4 count of HIV-infected patients as a function of Highly Active Antiretroviral Therapy initiation and other relevant baseline characteristics of the patients. Thus, even though this is a biostatistics methodological dissertation research, some interesting clinical and sociodemographic findings are also discussed. Discussion and conclusion of the results from the proposed models with a suggestion of possible further research avenues completed the study. Iqoqa Kule dizetheshini izindlelasu zokulinganisela imininingo eziyilongitudinal modeling approaches ukuhlaziya imininingo yezibalo zamasosha e-CD4 count ezikalwa ngokuphindaphindeka ezigulini ezitheleleke nge-HIV ezibhalise eSikhungweni soHlelo Lokucwaninga nge-AIDS eNingizimu Afrika kuyaphenywa. Imininingo enqumile, noma imininingo ekalwa ngokuphindelela, iwuhlobo oluqondile lwemininingo emazingeni ahlukene. Ocwaningweni olunqumile, ukubheka okuphindwayo kwenziwa kumuntu oyedwa emphumeleni owodwa noma engaphezulu, okufaka ulwazi lwamakhovariyenti njengesisekelo nangokuhamba kwesikhathi. Amamodeli anemithelela exubile aseyathandeka ekuhambiselaneni nemininingo yemodeli enqumile. Inqubo yezibalomininingo iphinde ivumele ukuqagula ukuguquguquka emininingweni enomumo onokugibelana iphinde ihlole imithelela yezimo emazingeni ehlukene. Njengoba ucwaningo olunqumile luvame ukubhekana nokungaphothulwa kwemininingo ngenxa yabantu ababhekwe ingxenye, imodeli enomthelelangxube ngokomumo wayo iyakwazi ukubhekana nemininingo engabhalansile eyilolu hlobo bese luhlonza izimo zokokusebenza kwengqondo nomumoqoqobantu emphakathini emininingweni ethinta i-CD4 count eqoqweni leziguli ezitheleleke nge-HIV. Njengoba kunentshisekelo enkulu ukuba umhlaziyi wezempilokwelapha noma umphenyi enze imodeli ngendlela ukubalwa kwamasosha i-CD4 count noma amabhayomakha esifo esigulini ebukhoneni bamakhovariyenti noma izimo ezihlonza ukuqhubeka kwesifo ngokuhamba kwesikhathi, indlelasu yokunqandeka kwesifo ngokukaPoisson, okuchaza ukuguquguquka ngokubalwa, kuyabhekwa. Amamodeli kaPoisson abekwe eceleni anemithelela exubile angaba wukukhetha okuyikho kwemininingo yokubala kokuphindelela. Kodwa, le modeli ayivezi okuyikho ngenxa yokuvimbeleka ukuthi imini nevariyenti kuyalingana. Ngakho-ke, imodeli kaPoisson enemithelelangxube imelwe yimodeli enemithelelangxube yebhayinomiyali engeyinhle. Imodeli yakamuva ilawula ngempumelelo yokusabalalisa kakhulu imininingo enqumile. Sihlaziya siphinde siqhathanise namamodeli aphakanyisiwe nokusetshenziswa kwayo ekubaleni amasosha omzimba i-CD4 cell count yeziguli ezitheleleke nge-HIV abafakwe ekubambeni iqhaza emininingweni yocwaningo. Imiphumela iveza ukuthi imodeli yemithelelangxube yebhayinomiyali engeyinhle enezakhiwomumo eyiwo nesebenza yedlule ekaPoisson nemodeli enomthelelangxube ngokwemigomo yokubheka ukusabalalisa ngokweqile imininingo. Amasu amaningi emvezabubi asetshenziselwa ukubhekana nezimo ezingabonakali kwisethi yemininingo ehlaziya iziqagulo zamapharamitha ekufakweni kwemodeli enemithelelangxube yebhayinomiyali ngokuthatha ngokuthi kunokungatholakali okungahleliwe. Ukukhombisa ukusabalalisa okugcwele okunemibandela komphumela ophindaphindekile, imodeli enemithelelangxube iyasetshenziswa. Lokhu kuveza ukufaka imodeli yezibalomininingo ezifaka konke okunamamodeli ajwayelekile ayingxube. Ukuncipha kwekhwantayli kunika ithuluzi elingenamsebenzi ukuhlonza imithelela ebingetholwe amamodeli ejwayelekile okuncipha, agxile kuphela kwimini encike ekufakweni kwemodeli. Imodeli yokuncipha kwekhwantayli evuma ukusabalalisa i-Laplace etshekile yetemu elingene ngephutha kwasetshenziswa emininingweni yokubala i-CD4. Ukuqagula okuyikho okuphezulu kwemithelela yamakhovariyenti nezakhi zekhovariyensi-variyensi kwimodeli enemithelelangxube yekhwantayli eyaqaliswa ukusebenza kusetshenziswa i-algorithimu i- Stochastic Approximation Expectation-Maximization. Kwimodeli, imithelela engahlelekile emininingo iphinde yafakwa ukuze kubhekwane nokuxhumana kokuqashelwayo. Ngakho-ke, sithola ukuqagula amapharamitha okunzulu ngemumo yokusabalalisa okunemigomo eyehlukahlukene okunika isithombe esifaka konke nesiphelele semithelela. Ngaphezu kwalokho, ukuthola imibono eyongeziwe ngobudlelwane obusebenzayo phakathi kwevariyebhuli yempendulo namakhovariyethi, amamodeli anemithelelangxube eyongezwayo, njengemodeli yemithelela exubile engemihle yebhayinomiyali eyongezwayo, imodeli enguqunguqu isetshenziselwe ukuqonda kangcono ngemodeli yokuhlaziya izinkombakusasa ezingenamigoqo ezinkimbi emininingweni engumumokuqonda ohlakazwe kakhulu, iyasetshenziswa. Uma kulandelwa imodeli enemithelelangxube yebhayinomiyali engeyinhle eyongezwayo, ukuze kuhambelane nomzamo wokufaka imodeli enemithelelangxube ayikhwayintali eyongezwayo, uhlaka oluguqulekayo nolusebenza ngendlela yezindlela ezingahambelani nepharamethrikhi kanjalo nepharamethrikhi engumumokuqonda wokuhlaziya imininingo okugxile ezicini zomphumela owedlula injwayelosenzo ewumongo, nakho kwenziwa. Ivariyebhuli yempendulo esebenzayo yisibalo samasosha omzimba i-CD4 ezigulini ezitheleleke nge-HIV njengomzamokuziqamba Wengxubekwelapha Ethithibalisa igciwane leSandulela Ngculazi Esebenza Kakhulu kanye nezinye izici eziyisisekelo eziyiso zeziguli. Ngakho-ke nakuba lena kuyidizetheshini yocwaningo lwendlelakwenza yocwaningo kwezibalomininingokuphila, kuphinde kwadingidwa okutholakele kwezempilongqondo nakwisifundomumoqoqobantu emphakathini. Ukudingida nokuphothulwe yimiphumela yamamodeli aphakanyisiwe ngesiphakamiso sezindlela zokukwenza ucwaningo oluqhubekayo ukuphothula ucwaningo.
Statistical models to study the BMI of under five children in Ethopia.
(2018) Yirga, Ashenafi Argaw.; Mwambi, Henry Godwell.; Ayele, Dawit Getnet.; Melesse, Sileshi Fanta.
Maternal and child malnutrition has long and short-term consequences on the health status of the people and on the country’s economy. It is among the major public health problems in Ethiopia. Worldwide, maternal and child malnutrition is an underlying cause for more than 3.5 million deaths each year. About 35% of the global disease burden is in under five children. Such a heavy burden requires an understanding of the nutritional status of the people, especially children under the age of five years and associated factors. Therefore, this study attempted to use possible statistical methods to estimate the effects of the risks related to the nutritional status of children. It also tried to identify the socio-economic and demographic factors that are associated with the BMI of under five children in Ethiopia. The study employed the 2016 Ethiopian Demographic and Health Survey data. A nationally representative sample of children under the age of five years was used to get information on weight and height measures of under five children. The BMI of children under five years of age was used as a response variable to fit weighted quantile regression. The covariates, age of a child, sex and other relevant socio-economic and demographic factors were used in the study. Following the quantile regression, the generalized linear models such as logistic regression model was applied after categorizing the response variable, BMI of under five children, into two categories namely normal and malnourished. Following binary logistic regression, an attempt to fit ordinal logistic regression was made. That means nutritional status was considered as ordinal outcome with four categories namely underweight, normal, overweight and obese. The findings and comparison of estimates using these different statistical methods with and without complex survey design were presented. The results revealed that methods that take into account the complex nature of the design, perform better than those that do not take this into account. It has also been found that age of a child, weight of child at birth, mother’s BMI, educational attainment of mother, region and wealth index were significantly associated with under five children’s nutritional status. Furthermore, the results are discussed and then a conclusion is made in the context of policy implication for Ethiopia.
Use of statistical modelling and analyses of malaria rapid diagnostic test outcome in Ethiopia.
(2013) Ayele, Dawit Getnet.; Zewotir, Temesgen Tenaw.; Mwambi, Henry G.
The transmission of malaria is among the leading public health problems in Ethiopia. From the total area of Ethiopia, more than 75% is malarious. Identifying the infectiousness of malaria by socio-economic, demographic and geographic risk factors based on the malaria rapid diagnosis test (RDT) survey results has several advantages for planning, monitoring and controlling, and eventual malaria eradication effort. Such a study requires thorough understanding of the diseases process and associated factors. However such studies are limited. Therefore, the aim of this study was to use different statistical tools suitable to identify socioeconomic, demographic and geographic risk factors of malaria based on the malaria rapid diagnosis test (RDT) survey results in Ethiopia. A total of 224 clusters of about 25 households were selected from the Amhara, Oromiya and Southern Nation Nationalities and People (SNNP) regions of Ethiopia. Accordingly, a number of binary response statistical analysis models were used. Multiple correspondence analysis was carried out to identify the association among socioeconomic, demographic and geographic factors. Moreover a number of binary response models such as survey logistic, GLMM, GLMM with spatial correlation, joint models and semi-parametric models were applied. To test and investigate how well the observed malaria RDT result, use of mosquito nets and use of indoor residual spray data fit the expectations of the model, Rasch model was used. The fitted models have their own strengths and weaknesses. Application of these models was carried out by analysing data on malaria RDT result. The data used in this study, which was conducted from December 2006 to January 2007 by The Carter Center, is from baseline malaria indicator survey in Amhara, Oromiya and Southern Nation Nationalities and People (SNNP) regions of Ethiopia. The correspondence analysis and survey logistic regression model was used to identify predictors which affect malaria RDT results. The effect of identified socioeconomic, demographic and geographic factors were subsequently explored by fitting a generalized linear mixed model (GLMM), i.e., to assess the covariance structures of the random components (to assess the association structure of the data). To examine whether the data displayed any spatial autocorrelation, i.e., whether surveys that are near in space have malaria prevalence or incidence that is similar to the surveys that are far apart, spatial statistics analysis was performed. This was done by introducing spatial autocorrelation structure in GLMM. Moreover, the customary two variables joint modelling approach was extended to three variables joint effect by exploring the joint effect of malaria RDT result, use of mosquito nets and indoor residual spray in the last twelve months. Assessing the association between these outcomes was also of interest. Furthermore, the relationships between the response and some confounding covariates may have unknown functional form. This led to proposing the use of semiparametric additive models which are less restrictive in their specification. Therefore, generalized additive mixed models were used to model the effect of age, family size, number of rooms per person, number of nets per person, altitude and number of months the room sprayed nonparametrically. The result from the study suggests that with the correct use of mosquito nets, indoor residual spraying and other preventative measures, coupled with factors such as the number of rooms in a house, are associated with a decrease in the incidence of malaria as determined by the RDT. However, the study also suggests that the poor are less likely to use these preventative measures to effectively counteract the spread of malaria. In order to determine whether or not the limited number of respondents had undue influence on the malaria RDT result, a Rasch model was used. The result shows that none of the responses had such influences. Therefore, application of the Rasch model has supported the viability of the total sixteen (socio-economic, demographic and geographic) items for measuring malaria RDT result, use of indoor residual spray and use of mosquito nets. From the analysis it can be seen that the scale shows high reliability. Hence, the result from Rasch model supports the analysis carried out in previous models.