DSpace Collection:http://41.89.96.81:8080/xmlui/handle/123456789/25362024-03-29T09:22:50Z2024-03-29T09:22:50ZMissing data imputation in multivariate distribution with unknown degrees of freedom using expectation maximization algorithm and its stochastic variantsKinyanjui, Paul KimaniTamba, Cox LwakaOrawo, Luke Akong’oOkenye, Justin Obwogehttp://41.89.96.81:8080/xmlui/handle/123456789/32422024-01-26T08:01:47Z2020-10-01T00:00:00ZTitle: Missing data imputation in multivariate distribution with unknown degrees of freedom using expectation maximization algorithm and its stochastic variants
Authors: Kinyanjui, Paul Kimani; Tamba, Cox Lwaka; Orawo, Luke Akong’o; Okenye, Justin Obwoge
Abstract: . Many researchers encounter the missing data problem. The phenomenon may be occasioned by data omission, nonresponse, death of respondents, recording errors, among others. It is important to find an appropriate data imputation technique
to fill in the missing positions. In this study, the Expectation Maximization (EM) algorithm and two of its stochastic variants,
stochastic EM (SEM) and Monte Carlo EM (MCEM), are employed in missing data imputation and parameter estimation in
multivariate t distribution with unknown degrees of freedom. The imputation efficiencies of the three methods are then compared
using mean square error (MSE) criterion. SEM yields the lowest MSE, making it the most efficient method in data imputation
when the data assumes the multivariate t distribution. The algorithm’s stochastic nature enables it to avoid local saddle points and
achieve global maxima; ultimately increasing its efficiency. The EM and MCEM techniques yield almost similar results. Large
sample draws in the MCEM’s E-step yield more or less the same results as the deterministic EM. In parameter estimation, it is
observed that the parameter estimates for EM and MCEM are relatively close to the simulated data’s maximum likelihood (ML)
estimates. This is not the case in SEM, owing to the random nature of the algorithm.
Keywords: Expectation maximization (EM), stochastic EM, Monte Carlo EM, unknown degrees of freedom2020-10-01T00:00:00ZEffect of Employee Demographic Characteristics and Family/Work Conflict on Organizational Commitment Among the Police Officers in Nakuru County, KenyaKipkebut, DinahLekaram, Estherhttp://41.89.96.81:8080/xmlui/handle/123456789/32402024-01-26T07:34:34Z2019-03-01T00:00:00ZTitle: Effect of Employee Demographic Characteristics and Family/Work Conflict on Organizational Commitment Among the Police Officers in Nakuru County, Kenya
Authors: Kipkebut, Dinah; Lekaram, Esther
Abstract: This paper examines the demographic characteristics and related variables on family/work conflict among the Police Officers in Nakuru County, Kenya. The participants were from eight (8) Police Division and a simple random sample of 170 was derived from a total number of 278 Officers. The level of conflict that employees perceive between their working and personal lives on organizational commitment was based on their personal characteristics which included: gender, marital status, age, level of education, tenure and number of hours worked among the police officers. The report suggested that the biographical personal characteristics suggested that those who participated there were more male officers compared to the female gender. The results showed that more female police officers experienced more stress than men which spill into their work due to the multiple roles they play as wives, mothers, care givers, family career as well as employees. The results are discussed in terms of marital status, level of education, tenure of service and numbers of hours worked per day on how they influence organizational commitment among the police officers in Nakuru County, Kenya.2019-03-01T00:00:00ZMangrove cover and cover change analysis in the trans boundary area of Kenya and Tanzania during 1986 2016Mungai, FredrickKairo, JamesMiringa, JohnKirui, BenardMangora, PeterKoedam, Nicohttp://41.89.96.81:8080/xmlui/handle/123456789/30192023-11-29T07:25:53Z2019-04-01T00:00:00ZTitle: Mangrove cover and cover change analysis in the trans boundary area of Kenya and Tanzania during 1986 2016
Authors: Mungai, Fredrick; Kairo, James; Miringa, John; Kirui, Benard; Mangora, Peter; Koedam, Nico
Abstract: Mangrove forests are among the most threatened ecosystems on earth. Some of these forests traverse national boundaries complicating their management due to differences in governance structures between countries. To improve the management of transboundary species regular monitoring is essential. Remotely sensed data were used to estimate forest cover and analyze conditions of mangroves in the proposed transboundary conservation area (TBCA) between Kenya and Tanzania. Image analysis was performed using unsupervised and supervised classification methods. The transboundary mangroves cover an estimated 11,906 ha; 55% being in Kenya, 45% in Tanzania. Ceriops tagal, Avicennia marina, and Rhizophora mucronata species co-dominate the mangroves of the transboundary area. The hotspot for loss and degradation of mangrove in the TBCA is Vanga in Kenya with a loss of 27 ha/yr. Harvesting of mangrove wood products have contributed to the loss of mangroves in the transboundary area. TBCA formation could play a critical role in ensuring sustainable mangrove resources utilizatio2019-04-01T00:00:00ZTraditional Oath Administration and Cleansing in Africa: The Case of the Akamba Ethnic Group in KenyaMuliro, Monica N.Theuri, Matthew M.Matheka, Reuben M.http://41.89.96.81:8080/xmlui/handle/123456789/27222021-08-10T08:51:15Z2015-06-01T00:00:00ZTitle: Traditional Oath Administration and Cleansing in Africa: The Case of the Akamba Ethnic Group in Kenya
Authors: Muliro, Monica N.; Theuri, Matthew M.; Matheka, Reuben M.
Abstract: This study examined traditional oath administration and cleansing among Akamba in Machakos County, Kenya. This study was conducted as a qualitative case study in Machakos Sub County of Machakos County. The target population was the Kamba Christians in Machakos County. Purposive sampling technique was used in sampling Kamba Christians from Mainline Churches and those who participated were selected using snowball sampling technique. Informants identified were subjected to interview schedules to provide primary data for the study. Primary data was collaborated and supplemented by secondary data collected from various libraries. The data collected was edited, coded, classified and analysed. This study learnt of the various ways Kamba oathing (Kithitu) is administered and cleansed like public performance by a religious
specialist, Ndundu oath, breaking of the earthen ware pot, swearing on the loin clothes by women, burning and urinating on a piece of wood, pouring millet on fire, and the use of soil in land conflict. However one can be cleansed of the effects of the oath. From the findings, this research concluded that Kithitu has a modern significance and it is interrelated and supported by continuity of other African traditional concepts like kinship ties, belief in ancestors, rites of passage, taboos and community. This research consequently recommended a further and contemporary study of African traditional Religious concepts as one main source of doing African Christian theology.
Keywords: African Traditional Religion, Traditional oathing, Cleansing, religious specialists2015-06-01T00:00:00Z