Statistics

Permanent URI for this collection

http://etd.aau.edu.et/handle/123456789/32

Browse

Browsing Statistics by Author "Abegaz Fentaw (PhD)"

Now showing 1 - 5 of 5
  • No Thumbnail Available
    Item
    Application of Garch Models in Forecasting the Volatility of Export Prices
    (Addis Abeba university, 2009-02) Terefe Amare; Abegaz Fentaw (PhD)
    This thesis discusses the GARCH model fitting and volatility forecasting of export prices data. We have chosen to confine our analysis on total export prices, coffee export prices and oil seeds export prices. The implied goal is to fit an appropriate GARCH model, to find out if the mentioned export prices are volatile or not and to forecast the volatility for some future times. ARMA(1,1) is found as the most appropriate model for the conditional mean of total, coffee and oilseeds export prices and it is also found that GARCH(2,1) for modeling volatility of total export prices and coffee export prices, and GARCH(2,2) for modeling volatility of oil seeds export prices as best models. Moreover, the results suggest that the export prices volatility is persistence in all the three cases indicating that past volatility is important in predicting (forecasting) future volatility. Key Words: export, GARCH, volatility, forecasting, Ethiopia
  • No Thumbnail Available
    Item
    Bayesion an Approach to Identify Predictors of Chilren Nutritional Status in Ethiopia
    (Addis Abeba university, 2009-06) Mesele Tesfaye; Abegaz Fentaw (PhD)
    Child Mortality and Undernutrition are the two major public health problems in the developing world. The associated phenomena are household living style, type of residence and awareness in health and family planning programs. Ethiopia, in particular, suffers from worst forms of malnutrition due to access to health care and nutrition. This thesis explores the most dominant socio economic, demographic and environmental factors in children nutritional status. We used recently developed Bayesian structured additive models to flexibly model the effects of included covariates on the EDHS 2005 datasets of Ethiopia. Inference is fully Bayesian based on recent Markov chain Monte Carlo techniques. These models allow us to analyze usual linear effects of categorical covariates and nonlinear effects of continuous covariates within a unified semiparametric Bayesian framework for modeling and inference. Most of the socioeconomic, demographic and environmental determinants included in the study were found to be statistically significant with the exception of covariates household economic status and mother education
  • No Thumbnail Available
    Item
    Predictors of Growth of Teledensity in Ethiopian Telecommunications Corporation
    (Addis Abeba university, 2009-06) Meressa Zewdu; Abegaz Fentaw (PhD)
    The Ethiopian Telecommunications Corporation is the sole provider of national and international telecommunications in Ethiopia since 1894. In Ethiopia, teledensity grew so slowly, as compared to developed countries or other developing countries. This study uses 31 years of yearly data starting from the beginning of 1970 in Ethiopian fiscal year. The problems and strategic actions for growth in teledensity are discussed. Also, the opportunities for utilizing information and communication technologies to solve priority problems and to realize sustainable development in the country are examined. Parametric (Cochrane‐Orcutt) and nonparametric (lowess) multiple regression models are employed. Specifically, the findings of the parametric regression model based on Cochran‐Orcutt transformation to handle serial correlation of residuals suggest that major determinant for growth of teledensity are higher GDPC and higher contribution of the service sector share to GDP in Ethiopia. And the average revenue generated by each telephone line (average used charge) is negatively related to teledensity in Ethiopia. In addition, the non parametric regression based on lowess method fitted the teledensity data equally as good as the parametric method
  • No Thumbnail Available
    Item
    Socio-Demographic Factors Affecting Bipolar Disorder Patients: the Case of Butajira, Ethiopia
    (Addis Abeba university, 2009-06) Zemicael Aklilu; Abegaz Fentaw (PhD)
    The purpose of this study is to identify socio-demographic factors affecting the bipolar disorder patients. The data included in this study are a cross sectional data collected by the Butajira mental health project. In this study, 828 psychiatric patients were included, of which 511(61.75%) belong to the non bipolar disorder symptom, 218(26.3%) belong to the moderate bipolar disorder symptom and 99(12%) belong to severe bipolar disorder symptom. In the analysis of data, polytomous logistic regression and multiple discriminant analysis are employed. It is found that place of residence, quality of life, marital status, number of children, and the interaction between residence and quality of life are significantly associated with the severity of bipolar disorder. The polytomous logistic regression provides higher percentage of correctly classifying psychiatric patients into no, moderate, and severe bipolar disorder symptom groups (61.7%) as compared to multiple discriminant analysis (50.8%). So that polytomous logistic regression is more efficient than that of multiple discriminant analysis in classifying of the psychiatric patients into the no, moderate and severe bipolar disorder symptom groups. This study identifies that the potential socio-demographic factors such as place of residence, quality of life, number of children, marital status, and age of the patients and the interaction between residence and quality of life affects bipolar disorder patients
  • No Thumbnail Available
    Item
    Survival Analysis of Time to Treatment Resumption for Chronic Hiv-1 Patients Interrupting Highly Active Antiretroviral Therapy (Heart)
    (Addis Abeba university, 2009-06) Anagaw A.Berhanu; Abegaz Fentaw (PhD)
    Highly active antiretroviral therapy (HAART) has significantly reduced mortality caused by human immuno-deficiency virus (HIV) by enhancing the physiological and immunological ability to counterattack against the virus and increases the life expectancy. Once started, the antiretroviral treatment should be continued lifelong and adherence to this treatment should be nearly perfect to enable long-term efficacy. A continuous and life-long treatment with HAART may lead to a broad spectrum of significant toxicities. As a result many patients interrupt their HAART without the knowledge and advice of the clinicians and this cause the patient immune to degrade and even cause to death. Therefore, this study is an attempt to examine the predictive factors of interruption duration in patients who interrupt their ARV drugs treatment. The study is a retrospective cohort study of HIV/AIDS patients under HAART but discontinued their treatment for at least one month, which comprises a total of 723 patients from Zewditu Memorial Hospital. Of these patients about 67% resumed their treatment. In the analysis of data the Kaplan-Meier survival estimator and Cox proportional hazards regression model are used. Based on the data analysis, it is found that main effects: education, baseline age, baseline weight, CD4 count at the start of HAART and prior to interruption, duration on HAART and interaction effects: marital status with disease duration and marital status with sex are important factors that are related to time to resumption of HAART. In general, HIV/AIDS patients who have prolonged treatment interruption are characterized by having lower level of education, younger age, high weight, higher CD4 count at the start of HAART, lower CD4 count prior to treatment interruption, longer duration on treatment follow-up and currently not married and male patients with longer disease duration