# STATISTICS

## On the Inverse Burr Distribution: Its Properties and Applications. Journal of the Nigerian Association of Mathematical Physics. Vol. 42, pp. 61 – 66.

This paper presents a discussion on the mathematical properties of the inverse Burr distribution.  The application of the distribution was subjected to two lifetime data sets and some measures of goodness-of-fit such as the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Komolgorov-Smirnov (K-S) test statistics were considered to examine its flexibility in modeling lifetime data and the superiority over other well-known distributions.  Results obtained from the two lifetime data sets, reveal that the inverse Burr distribution is an appropriate model in fitting

## An Appraisal on Some Methods for Estimating the 2-Parameter Weibull Distribution with Application to Wind Speeds Sample. Sri Lankan Journal of Applied statistics, Vol. 19(3).

Six methods for estimating the Weibull shape and scale parameters are considered and compared in this paper.  These methods are: the least squares method, weighted least squares method, method of moments, energy pattern factor method, method of L-moments and the maximum likelihood method.  A simulation study as well as application to a real data set (wind speeds sample) was used to test the performance of different methods using the smallest mean square error criterion.  Results from the simulation study indicated that the maximum likelihood method is the most efficient method when dealing

## A New Approach for Improving Classification Accuracy in Predictive Discriminant Analysis. Annals of Data Science. Vol. 5(3), pp. 339357.

The focus of a predictive discriminant analysis is to improve classification accuracy, and to obtain statistically optimal classification accuracy or hit rate is still a challenge due to the inherent variability of most real life dataset.  Improving classification accuracy is usually achieved with best subset of relevant predictors obtained by using classical variable selection methods.  The goal of variable selection methods is to choose the best subset (or training sample) of relevant variables that typically reduces the complexity of a model and makes it easier to interpret, improves t

## On the Use of Predictive Discriminant Analysis in Academic Prediction. Journal of the Nigerian Statistical Association. Vol. 29, pp 71 – 80.

Selection of key predictor variables in classical statistical procedures such as predictive discriminant analysis (PDA) not only leads to identification of key predictor variables which separate the groups well, but also improves prediction or classification accuracy.  Because the dependent variable in PDA is categorical, the technique lends itself to various uses in higher education predictions.  But the predictive validity of the predictive discriminant function (PDF), in the context of academic prediction can best be evaluated based on the relevance of the selected key predictor variable

## Statistical Association, Vol. 29, pp. 47 – 57

A Modified Intersection of Confidence Intervals Approach in the Multivariate data density estimation.  Journal of the Nigerian .

## On Maintaining High Levels of Immunization Coverage in Edo State Using Binary Logistic Regression Model. Journal of the Nigerian Association of Mathematical Physics. Vol. 35, pp 201 – 208. A target of 90-95% levels of immunization coverage is necessary f

In Nigeria, achieving and maintaining a high level of 90-95% immunization coverage which is an indication of the effectiveness of a health care system has remained elusive till date.  This paper is therefore aimed at exploring the possibility of maintaining high levels of immunization coverage in Edo state using best subset binary logistic regression model and the chi-squared test as validation tool.  We found that the obtained logistic model which constitutes only five key predictor variables out of eight potential predictors used had a good predictive performance.  In addition, validati

## A Comparative Study of K-Means and K-Medoids Clustering Methods. Journal of the Nigerian Association of Mathematical Physics. Vol. 36(2) pp 169 176.

The aim of this work is to provide a formal and organized study of the effect of the nature of data and cluster structure on the performance of K-means and K-medoids clustering methods.

## Hybrid of ARIMA-ARCH Modelling of Daily Share Price Data of Okomu Oil Plc in Nigeria. Journal of the Nigerian Association of Mathematical Physics. Vol. 36(2) pp. 163 – 168.

The aim of this work is to study and develop an appropriate time series model for the residuals from the autoregressive integrated moving average (ARIMA) model derived from the daily stock data of Okomu Oil.

## A New Class of Winsorized Shrinkage Estimators for Multiple Linear Regression. Transactions of the Nigerian Association of Mathematical Physics. Vol. 2 pp 143 – 150.

In this paper, the simultaneous occurrence of multicollinearity and legitimate contaminant in Y-space due to non-normality of error variable is considered.  To handle the problem of multicollinearity and legitimate contaminants in the data, a new class of modified Winsorized shrinkage estimators (MWSEs) is proposed and their performance is evaluated through estimated mean square error (EMSE) sense.  A simulation studies reveal that the MWSEs show consistently minimum EMSE among the considered shrinkage estimators.

## Using Discriminant Analysis to Identify Major Prerequisites for Success in Specific Courses of Study in a University system. Journal of the Nigerian Association of Mathematical Physics. Vol. 33, pp 99 – 106.

Inappropriate choice of course of study at tertiary institutions results in unacceptable levels of attrition and poor throughput rate.  The impact on students can cause lasting damage to selfesteem and consequences can influence an entire lifetime.  In this paper discriminant analysis was used to identify major prerequisite for success in a specific course of study in a university system.  We present a case study with forty (40) industrial mathematics majors where discriminant analysis was used successfully to determine the major prerequisite for success in industrial mathematics.