The article aimed at fitting Cox-proportional hazards model to Tuberculosis (TB) data. TB data on 259 patients spanning 2010 through 2016 were collected from the Federal Medical Centre, Bida, Nigeria. Covariates involved were gender, age, type of TB and occupation. Fifteen different Cox models, representing all possible combinations of covariates in question were fitted. Parameters were estimated by method of maximum partial likelihood and model selection was based on Akaike information criterion (AIC). Model (G+C), with gender and occupation as covariates produced the least AIC of 618.597 and hence, was adjudged the best. That is, gender and occupation constituted the best subset of covariates that explained survival of TB patients. The model suggested that recovery hazard of a male TB patient was 24.1% lower than that of a female patient possessing same occupation. This implies that male patient had higher survival time than the female having same occupational status. It further suggested that recovery hazard for patient on technical occupation was 27.46% higher than for patient on non-technical job and of same gender. Hence, a patient on technical occupation had reduced survival time compared to one of same gender on non-technical occupation. It was concluded that gender and occupation explained best, survival of TB patients based on AIC.

Survival analysis is a class of statistical method aimed at studying the occurrence and timing of events of interest. The analysis is propelled by event while time is the central operator. In medicine, event may be death, relapse, in which wise, time may be time to death and time to relapse respectively. In economics, the event may be employment while time is time to employment, that is, unemployment duration. In Civil Engineering, the event may be completion of a project while time is project duration; the event may also be appearance of a crack on a building while the time istime that lapses between project completion and appearance of the crack. It is thus, a concept applicable to virtually all aspects of human endeavour where event and time to event are clearly defined. Unemployment duration study is very important as it explains changes in the labour market situation. Such a study is capable of informing policy makers about effectiveness of policies formulated to tackle unemployment in an economy. An unemployment duration study via survival analytical approach that signifies that unemployment duration has reduced significantly will naturally be pleasing to the government. A contrary result will suggest that government approach to tackling unemployment needs to me revisited. The need for such study from time to time can therefore, not be overemphasized. A large amount of efforts has gone into survival analysis research. Although most of the efforts seem to be in the medical sciences, some applications have been made in other areas also. Leo and Go (1977) reviewed the common statistical technique employed to analyze survival data in public research; Tatsiramos (2006)studied the effect of unemployment insurance on unemployment duration and the subsequent employment stability using mixed proportional hazard model; Blanchard and Diamond (2008) examined unemployment duration dependence and suggested that this affects both the matching and the wage function. Daniela-Emanuela and Cirnu (2014) studied unemployment duration in Romania using survival methods; Novella and Duvivier (2015) examined the relationship between unemployment duration and education in Belgium;Oujezsky, Horvath, and Skorpil (2016) applied survival methods to analyze botnet command and control traffic using Kaplan-Meier estimator. Echeburua, Gomez and Freixa (2017) examined schizophrenia patients with gambling disorder using Cox survival model. These research objectives are to apply Kaplan-Meier survival model to unemployment duration data and to also compare the unemployment duration experience of males and females. The remaining part of the article is organized as follows: Section 2 presents Methodology; Section 3 presents Results and Discussion while Section 4 concludes the article and recommends.

The research performed parametric survival analysis of Tuberculosis (TB) data (covering 2010 to 2016) collected from the Federal Medical Centre, Bida, Niger State, Nigeria. Three parametric survival models (Exponential, Weibull and Log-logistic) were fitted. The outcome variable was time to recovery from TB infection and four covariates being age, gender, TB type and occupation were involved. Models were estimated by maximum likelihood method and model selection criterion used was the Akaike Information Criterion (AIC). The exponential and log-logistic models found all covariates statistically insignificant while Weibull found all covariates but TB type significant at 5% level. Based on AIC, Weibull model with AIC of 163.5731 performed best, followed by log-logistic model with AIC of 191.419 and exponential model performed worst, with AIC of 517.9652. The best of fitted models being Weibull suggested that older patients had higher hazards than younger ones, older patients hence, had lower survival times, holding other covariates constant. That is, the older the TB patient, the lower was the time to recovery from TB. Males had higher hazards and hence, lower survival times compared to females. That is, male TB patients recovered faster than the females. Pulmonary TB patients had lower (insignificant) hazards and hence, higher survival times than Respiratory TB patients. TB patients on technical occupation had lower hazards than others and hence, had higher survival times than those whose occupations were considered not technical. The research concluded that age, gender and occupation were the major determinants of recovery period of TB patients. It was recommended that the Management of Federal Medical Centre, Bida, and other organizations involved in TB management could make use of the Weibull model to fit and predict both the survival and hazard rates of TB patients.

The order of bias of the fixed effects gompertz model is studied, using Monte Carlo approach. Performance criteria are bias and root mean squared errors. For fixed N, bias is found to decrease steadily between T=5 and T=20 but exhibits a mixture of increase and decline afterwards. At each value of T involved, bias steadily decreases with increased value of N. Bias is found to be at most 123%, due to the combination of minimum of each of N and T involved. Decrease in order of bias is found to be more definite with increased N at fixed T than with increased T at fixed N.

This article compared single to combined forecasts of wind run using artificial neural networks, decomposition, Holt-Winters’ and SARIMA models. The artificial neural networks utilized the feedback framework while decomposition and Holt-Winters’ approaches utilized their multiplicative versions. Holt-Winters’ performed best of single models but ranked fourth, of all fifteen models (single and combined). The combination of decomposition and Holt-Winters’ models ranked best of all two-model combinations and second of all models. Combination of decomposition, Holt-Winters’ and SARIMA performed best of three-model combinations and ranked first, of all models. The only combination of four models ranked third of all models. The accuracy of single forecast should not be underestimated as a single model (Holt-Winters’) outperformed eleven combined models. It is therefore, evident that inclusion of additional model forecast does not necessarily improve combined forecast accuracy. In any modeling situation, single and combined forecasts should be allowed to compete.