Statistics

Health- Related Quality of Life (HRQoL) is a notion that is commonly used to examine the impact of health status on quality of life. HRQoL is measured by the physical and psychosocial health quality of life of children and adolescent in this study. These two measures were used to provide a better picture of overall HRQoL of children and adolescent as individual measure may not capture the HRQoL as expected. To tackle this problem, this study developed a method for combining these two measures into a summary score called absolute HRQoL of children for ease of investigating the impact of drugs (first and second line ART) on the HRQoL of children/adolescents living with HIV/AIDS in Lagos State. The HRQoL data were verified to know whether the assumptions of normally distributed data and equality of standard deviations in the two groups were reasonably satisfied. These assumptions were not met due to skewness, orthogonal array-based Latin hypercube designs [OA (121,3) LHD and OA (16, 8) LHD] were adopted and the data was scaled to normally distributed data using a universally accepted mathematical formula. The standard deviations also showed that the largest standard deviation (20.53) is not more than twice as large as the smallest (15.91) and thus the assumption of equality of variances has been satisfactorily met which allowed the use of analysis of variance (ANOVA) approach on HRQoL among children and adolescent in Lagos State. The aim of this study is to determine the influence of the ART based on the first and second line drugs on the health related quality of life of children/adolescent living with HIV in Lagos State and determine the reliability of the result obtained in this respect. Y (Health Related Quality of Life) denotes the response variable and X represents the independent variable (first and second line drugs). The p-value obtained is greater than 0.05 indicating that there is sufficient evidence that the mean of the health related quality of life of children and adolescent on first line ART and those on second line ART do not differ significantly as shown in Table 6. Conclusively, children and adolescent in both groups are doing fine in terms of their health related quality of life. The rationale behind this may be attributed to their level of adherence to taking drugs at regular times and as prescribed.

The present paper examines levels and trends of fertility decline in Uttar Pradesh, which has occurred over time. The two major factors viz. wanted fertility and unwanted fertility have been the theme of several discussions over decades, among other aspects of population policy. Wanted births are those which are planned by couples and accepted socially, whereas unwanted births are those which do not adhere to certain social norms and may be categorized as illegitimate births, births after achieving a desired family size, births after marriage of own children etc. The study of unwanted pregnancy has been essential to demographers and public health practitioners to understand the levels of fertility and prevent unwanted childbearing. However, still a large number of the Indian women are forced to give unwanted births and this can be observed from the reports of NFHS-1, 2 & 3. Rural women in Uttar Pradesh are still having more children than their urban counterparts and the unwanted fertility rate is also higher. The fertility differentials exist by residence, education, religion, and caste/tribe in Uttar Pradesh. Research on wanted and unwanted components of fertility can lead to frame important policies which in turn may bring considerable decline in fertility and enhance the well-being of mothers and children. This paper measures the unwanted fertility by comparing total wanted fertility rate with the total fertility rate in Uttar Pradesh, which will ultimately help policymakers in designing proper action plan so that replacement level fertility can be achieved. Keywords: Fertility, TFR, TWFR, TUFR, NFHS, WFS, DHS

Cumulative Grade point average (CGPA) is a commonly used indicator for academic performance among students in the Universities. Many factors could act as barriers to students in attaining and maintaining a high CGPA that reflects their overall academic performance during their stay in the University. Such factors include test anxiety, time management, test competence, academic competence, study techniques, Lecturer teaching method, peer group, health, learning environment, parental background etc. In this study, a sequential procedure is used to model a combination of daily average reading hour and daily average sleeping hour to maximize students’ CGPA. The design for this study is stratified sampling scheme with department as the ultimate sampling unit. A Sample of 200 students was randomly selected. The questionnaire for the research contained information on students’ CGPA, daily average reading hour and daily average sleeping hour. SPSS software package was used to produce summary statistics. A second order response surface model (SORSM) was constructed from the data to determine various combinations of daily average reading hour and daily average sleeping hour that will make students to be in either good academic standing or not in good academic standing. The above SORSM was subjected to a canonical analysis to characterize the nature of stationary point of the model as well as to determine the daily average reading hour (x1) and daily average sleeping hour (x2) that maximize students’ CGPA. The model gives daily average reading hour (x1) to be 4 and daily average sleeping hour (x2) to be 5. Hence, the student’s maximum CGPA would be achieved if daily average reading hour (x1) is four and daily average sleeping hour (x2) is five after normal daily lectures.

In this paper we use 2-designs whose incidence matrix can take only binary values and construct a series of balanced and partially balanced n-ary t-designs by using a tool Block Sum and Product (BSP) Methodology. The simple 2- (6, 3, 2) design use as a parent 2-design for the procedure. This parent 2-design gives 36 new balanced n-ary t-designs and 3 partially balanced n-ary t-designs. We given incidence matrices and list out the parameters of newly constructed balanced and partially balanced n-ary t-designs.

The logistic regression model has had a remarkable growth in the field of statistics. It is particularly useful for predicting continuous and categorical explanatory variables using categorical dependent variable. This work studies 111 cases of HIV/AIDS infected patients and fits a binary logistic model. The results show that presence of tuberculosis, CD4 cell count and Regimens are significant risk factors of mortality of HIV/AIDS patients, obtained from the records Department of the General Hospital Wukari, Taraba State of Nigeria.

The optimality of the variance and efficiency balanced affine resolvable designs and resolvable designs with unequal block sizes has been checked and found it to be universal optimal. A method of constructing variance and efficiency balanced (?1, ?2, …..?t) resolvable balanced incomplete block design with unequal block sizes is also proposed using 2n-symmetrical factorial designs. Further optimality of the constructed design has been checked and found it to be universally optimal.

The relationship between crude oil prices and stock market indices has always been discordant. The article examines the performance of stock market with the help of different financial ratios used in oil and natural gas sector. Seventeen distinct companies involved in S&P BSE index are taken for our study. Initially 39 financial ratios are taken for analysis and a normality test was conducted on all these explanatory variables to find whether these variables are normal or not? A logistic regression was constructed with help of the selected nine financial ratios as independent variables for determining the performance of S&P BSE index as dependent variable. The classification result showed that, about 75.3 % accurate result for prediction ‘GOOD’ as well as ‘NOT GOOD’ performance of stock. Hosmer-Lemeshow test is applied to find out the overall model evaluation. The study also showed that logistic regression can be used by investors, fund’s managers, investment companies and researchers for academic interest as well as policy implications.

This study offers a brief overview of the export and import of wood based panelin India. The statistical parameters such as descriptive analysis, trends, graph and regression analysis have been computed to understand the nature of wood based panels export and import with top 25 countries on the basis of five year assessment, which is accounts 87.7% (1997-1998 to 2001-2002), 89.5% (2002-2003 to 2006-2007) and grow to 91% (2007-2008 to 2011-2012). The result of regression analysis demonstrate high R20.86 in export and 0.97 in import with correlated GDP, FDI and PC-NDP. The wood based panel’s viz. veneers, plywood, fibre boards, particle boards have also been examined for their trends.

Survival analysis as a field of statistics has had a remarkable growth in the latter half of the last century. The methods used by survival analysis that have led to this insightful impact are the Kaplan-Meier method used in estimating the survival function, the log-rank test commonly used to determine the differences between two or more survival distributions and the Cox Proportional Hazard model used for investigating the effects of covariates on the hazard function. The parametric methods are rarely used. In this work, we present the basic concepts of survival analysis. Furthermore, we compare the semi-parametric (Cox Proportional Hazard) and parametric methods for analysing survival data. We apply these methods to HIV/AIDS infected patients from secondary data to determine the prognostic factors of the event of mortality of the patients. It was found that the parametric method of survival analysis is good and probably preferred if the correct distribution is identified; and that drug abuse history of patients, tuberculosis status, patients’ adherence to drugs and the patients’ CD4 cell counts are prognostic factors for mortality experience of HIV/AIDS patients.

The present paper deals with the problem of estimating the finite population variance using two-phase sampling scheme in the presence of random non-response. In this paper, we have suggested some families of factor-type estimators of population variance utilizing the information on an auxiliary variable with unknown population variance. The properties of the suggested families of estimators have been discussed in detail. The optimum estimators of the suggested families have also been pioneered out. The theoretical results have been demonstrated through some real data sets. A simulation study has also been carried out to support the theoretical results.