Browsing by Author "Oyeyemi, G. M."
Now showing 1 - 20 of 66
Results Per Page
Sort Options
Item ALTERNATIVE ESTIMATOR FOR MULTIVARIATE LOCATION AND SCATTER MATRIX IN THE PRESENCE OF OUTLIER(Anale Seria Informatica, 2018) Obafemi, O. S.; Oyeyemi, G. M.It is generally known that in estimating location and scatter matrix of multivariate data when outliers are presents, the method of classical is not robust. The Maximum Likelihood Estimator (MLE) is always very sensitive to some deviations from the assumptions made on the data, especially, presence of outliers. To get over the above stated problem, many alternative estimators that are robust have been proposed in the last decades. Some of these estimators include the Minimum Covariance Determinant (MCD), the Minimum Volume Ellipsoid (MVE), S-Estimators, M-Estimators and Minimum Regularized Covariance Determinant (MRCD) among others. All the methods converged on tackling the problem of robust estimation by finding a sufficiently large subset of the data. In this paper, a robust method of estimating multivariate location and scatter matrix in the presence of outliers is proposed. The proposed estimator is obtained using the best units (samples) from the available data set that satisfied a set of three optimality criteria (CA,CH,CG).The performance of the proposed robust method was compared with two of the existing robust methods (MCD and MVE) and the classical method with their application in Principal component analysis data simulation. The measure of performance used was the Mean Square Errors (MSE) of the characteristic roots (eigen-values) of the variance covariance matrix. Generally, the proposed alternative method is better than other robust methods and classical method, when the level of magnitude of outliers is small and also performed considerably well with MCD and MVE when the level of magnitude is high at all percentages of outliers.Item An Alternative Solution to Hotelling T Square under the Heteroscedasticity of the Dispersion Matrix(Akamai University, U.S.A, 2019) Adebayo, P. O.; Oyeyemi, G. M.This work focuses on developing an alternative procedure to multivariate Behrens–Fisher problem by using approximate degree of freedom test which was adopted from Satterthwaite univariate procedure. The proposed procedure was compared via R package simulation and real-life data used by Timm with six (6) existing procedures namely: Johanson, Yao, Krishnamoorthy, Hotelling T square, Nel and Van der Merwe, and Yanagihara. It was discovered that the proposed procedure performed better in terms of power of the test, then all existing procedures considered in all the scenarios that include at different: (i) random variables (p), (ii) variance co–variance matrix, (iii) sample size, and (iv) significance level (α). The proposed procedure is completely favorably, well in terms of type I error rate with Johanson, Yao, Krishnamoorthy, Nel, and Van der MerweItem Analyzing Factorial Experiment Involving Qualitative and Quantitative Factors.(African Network of Scientific and Technology Institutions, Kenya., 2009) Oyeyemi, G. M.; Ibraheem, B. A.; Obafemi, O. S.; Ige, O. S.In factorial experiment involving qualitative and quantitative factors, ascertaining whether there is interaction between the two factors should be the first objective in the analysis. In this paper, we explore the different possibilities that may arise when dealing with qualitative and quantitative factors in a factorial experiment. The response curves for the quantitative factor are different across the levels of the qualitative factor when there is interaction between the factors, the curves are parallel but similar with distance apart when there is no interaction but both factors are significant, while curves converged to a curve (become a single curve) when the quantitative factor is not significant.Item Awareness of Highway Code Among Commercial Motorcyclists in Samaru, Nigeria(Nigerian Institute of Transport Technology (NITT), Kaduna, Nigeria, 2013) Arosanyin, G. T.; Olowosulu, A. T.; Oyeyemi, G. M.This paper examines Highway Code awareness among commercial motorcyclists in Nigeria. The study uses data generated from commercial motorcycle operators in Samaru, Zaria in Nigeria and; employed the logistic regression model to evaluate Highway Code awareness. It was found that about 41% of the operators are not aware of the existence of the booklet containing the Highway Code. The odds of Highway Code awareness increased with licence holding and experience but decrease with union membership and crash involvement. The Federal Road Safety Commission (FRSC) and Vehicle Licensing Office should enforce the requirements of theory and practical tests often based on the Highway Code to increase awareness and; organized seminars and training for the operators in partnership with the Motorcycle Union using the local language.Item Building A Quality Improvement Framework in the Transportation Industry(Mathematical Association of Nigeria, 2007) Oyeyemi, G. M.; Osanaiye, P. A.Not much empirical research works linking quality assessment and improvement programs exist to date for service industry and particularly for the transportation industry. The lack of such linkages has then resulted in firms using initiatives in a piecemeal manner and misunderstanding their impact, that is their potential or long-term impact. This paper is aimed at identifying quality improvement practices in the transportation industry in Kwara, Osun and Oyo States, categorizing them and linking them with performance outcomes. To achieve this, primary data on quality practices and measures of performance in the transportation industry was collected through administration of questionnaire for Kwara, Osun and Oyo states. Groups of Quality Constructs (Initiatives) were formed using factor analysis and Success of Transportation Quality Programs (STPQ) regressed on the Quality Constructs identified. The result showed that five Quality Constructs were paramount Quality Practices in the industry of which four are found to be significant to Success of the Transportation Quality Program (STPQ).Item Classification into Two Groups with different Cost of Misclassification Ratios(Nigerian Association of Mathematical Physics, 2016) Oyeyemi, G. M.; Oyebanji, L. A.Fisher’s Linear Discriminant and Bayesian Classification procedures were compared when the assumption of equal cost of misclassification is violated. The comparison was carried out at various sample sizes and different misclassification cost ratios. Data were simulated to consist two groups (populations) of four variables each from two multivariate normal populations. The homogeneity of the variance-covariance matrices of the two groups was tested using Box’s M-Test. The apparent Error Rate as the estimate of the Actual Error Rate was used to judge the performance of both procedures at different misclassification cost ratios (1:1, 1:2, , , 4:5) and sample sizes (10, 20, 30, 40, , , 100). The results show that at equal cost ratio (1:1), both approaches produced almost the same error rate at different sample sizes. With difference in misclassification cost ratio, the Bayesian approach generally has higher proportion of misclassification than the Fisher at various ratios and sample sizes. The Fisher performed better in small sample cases (n < 50) under all cost ratios considered except 1:2 and 1:5. For large sample cases (n > 50), the performance was better at cost ratios 2:3, 2:4 and 2:5Item Combined Exponential-Type Estimators for Finite Population Mean in Two-Phase Sampling(Asian Journal of Probability and Statistics, 2023) Oyeyemi, G. M.; Muhammad, I.; Kareem, A. O.A family of exponential-type estimator for estimating population mean in two-phase sampling when the population proportion of the auxiliary character is available is proposed in this paper. Theoretically, the bias and minimum mean square error (MSE) for the proposed estimator are obtained. The expression for MSE of the proposed exponential-type of estimator is compared with the existing estimators in the literature. The optimum values of the parameters are determined. An empirical study was carried out by comparing the proposed estimators with some of the existing estimators reviewed in the literature based on the criteria of bias, mean square error (MSE) and relative efficiency using life datasets. The result of the comparisons showed that the proposed exponential-type estimators produce a better estimate of finite population mean than the existing estimators in the sense of having higher percentage relative efficiency which implies lesser mean square error and bias. Furthermore, the realistic conditions under which the proposed class of exponential-type estimators is more efficient were also presented. Thus, the proposed estimators can be considered as significant alternatives to estimating population characteristics of real life datasets.Item Comparison between Fisherian and Bayesian Approach to Classification Using two Groups(College of Natural and Applied Science, University of Porthacourt, Nigeria, 2014) Oyeyemi, G. M.; Oyebanji, L. A.; Salau, I. S.; Folorunsho, A. ITwo approaches to discriminant analysis procedure are examined and compared based on their misclassification error rate. The Fisher's approach tends to find a linear combination of the variables which maximize the ratio of the between group sum of squares to that of the within group sum of squares in achieving a good separation. On the other hand, the Bayesian approach assigns an observed unit to a group with the greatest posterior probability. Fisher's linear discriminant analysis though is the most widely used method of classification because of its simplicity, and optimality properties is normally used for two group cases. However, Bayesian approach is found to be better than Fisher's approach because of its low misclassification error rate.Item Comparison of Bootstrap and Jackknife Methods of Re-sampling in Estimating Population Parameters(Published by Bachudo Sciences Co. Ltd, 2008) Oyeyemi, G. M.Re-sampling methods have been found to be useful for several purposes such as model selection linear regression and estimation of sampling variances or standard errors and confidence intervals. ln estimating the population coefficient of variation and its standard error, two methods of re-sampling Bootstrap and Jackknife are compared in this paper. The Jackknife method is found to require relatively small sample size to attain consistency in its estimate while Bootstrap requires larges ample size. Bootstrap is also found to always underestimate the standard error of its estimates.Item Comparison of Outlier Detection Procedures in Multiple Linear Regressions(Scientific and Academic Publishing, 2015) Oyeyemi, G. M.; Bukoye, A.; Akeyede, I.Regression analysis has become one of most widely used statistical tools for analyzing multifactor data. It is appealing because it provides a conceptually simple method for investigating functional relationship among variables. A relationship is expressed in the form of an equation or a model connecting the response or dependent variable and one or more explanatory or predictor variables. The major problem that statisticians have been confronted with, while dealing with regression analysis, is presence of outliers in data. An outlier is an observation that lies outside the overall pattern of a distribution. In other words it is a point which falls more than 1.5 times the interquartile range above the third quartile or below the first quartile. Several statistics are available to detect whether or not outlier(s) are present in data. Therefore, in this study, a simulation study was conducted to investigate the performance of Deffits, Cooks distance and Mahalanobis distance at different proportion of outliers (10%, 20% and 30%) and for various sample sizes (10, 30 and 100) in first, second or both independent variables. The data were generated using R software from normal distribution while the outliers were from uniform distribution. Findings: For small and medium sample sizes and at 10% level of outliers, Mahalanobis distance should be employed for her accuracy of detection of outliers. For small, medium and large sample size with higher percentage of outliers, Deffits should be employed. For small, medium and large sample sizes, Deffits should be used in detecting outlier signal irrespective of the percentage levels of outliers in the data set. For small sample and low percent of outliers Mahalanobis distance should be employed for easy computation.Item A Comparison of some Discriminant Techniques in Predicting Blood Pressure(Faculty of Science, University of Ibadan, Nigeria, 2019) Yussuf, T; Adeleke, B. L.; Oyeyemi, G. M.; Adeleke, M. O.; Kareem, A. O.The paper compared the performance of the Logistic Regression (LR), Fishers Discriminant Analysis (FDA) and the Support Vector Machines (SVM) in predicting high blood pressure. The variables used in the study are Age, Gender, BMI, Cholesterol and the smoking status. The SVM model was tuned to get the best parameter combination and cost function to avoid over fitting and under fitting of the model. The tuning model was used to compare the performance of the SVM model for sample sizes of 100, 500, 5000 and 7900. This approach was carried out for both the LR and FDA. The Data was divided into train and test data sets in the ratio 80:20 for all sample sizes considered to test the performance of the fitted models. The results from the sample sizes considered showed that for sample size of 100, the FDA performed better than the LR and SVM. But for sample sizes of 500, 5000 and 7900, the SVM performed better than both the LR and LDA. The area under the receiver operating curve showed 81.6% for the test data. This means that about 81.6% of the dataset was correctly predicted. The confusion matrix for the three approaches was computed. The result obtained showed the superiority of SVM to the other two methods.Item Comparison of some Robust Estimators in Multiple Regression in the Presence of Outliers(Akamai University, U.S.A, 2021) Oyeyemi, G. M.; Aji, D. A.; Ibraheem, B. A.; Kareem, A. O.Outlier results are one of the problems of Ordinary Least Squares (OLS) in regression analysis. Some estimators have been suggested as alternatives to the Ordinary Least Squares (OLS) estimator to improve the accuracy of the parameter estimates in the linear regression model in the presence of outliers. In this study, six robust estimators of handling the problem of outliers: Robust-M, Robust-MM; Robust-S; Least Trimmed Squares (LTS); Least Median Squares (LMS); and Least Absolute Deviation (LAD) were compared with OLS using Variance criterion. The multiple linear regression model considered, had 4 predictor variables (p = 5) and one dependent variable and there were four levels each of percentage of outliers (10%, 20%, 30%, 40%), variance of outliers (σ_outlier^2=1,50,100,200) and sample sizes (n = 20, 50, 100, 200) were considered through Monte Carlo experiments. The experiment was carried out 1000 times. The results showed that when the variance of outlier is 1, that is, the outliers and variables have standard normal distribution, OLS had the least variance at all sample sizes. But as the variance increases and at all sample sizes, the robust estimators outperformed the OLS. The robust MM had least variance more consistently as the sample size increases at all variance level of the outlier and also as the sample size increases. Therefore, the Robust MM Estimator performed more consistently than the other robust estimators considered.Item Comparison of Some Spike-and-Slab Priors for Bayesian Variable Selection in Multiple Linear Regression(Akamai University, U.S.A, 2019) Oyeyemi, G. M.; Olanrewaju, Y. A.; Kolawole, R. O.Variable selection has been a very essential challenge in building a multiple regression model. Exclusion of influential covariates or including covariate with zero effect will no doubt affect the estimation precision and as well the predictive accuracy of the model. “Spike-and-Slab prior” is an increasingly popular variable selection approach used in the Bayesian framework, which aids the variable selection and the estimation of regression parameters. In this research, the performances of the MCMC implementation for some versions of spike and slab priors for variable selection in normal linear regression models were investigated with regards to posterior inclusion probability for the simulated data under different setting (independent and correlated covariates, difference variance scales and varying sample sizes). Evidence from the simulation study revealed that the selected priors have similar performance under the independent setup and correlated setup, but the standard errors of coefficient estimates are higher for correlated covariates compare to independent covariates. The mean estimates of the coefficients get closer to the true coefficient values as the sample size increases under all different priors considered, and also the posterior inclusion probability depends on the size of variance of the slab component.Item Comparison of Taguchi Method of Analyzing Robust Parameter Design and Graphical Approach(Published by Faculty of Science, University of Ilorin, 2006) Oyeyemi, G. M.Taguchi's method/approach of analyzing data from Robust Parameter Design (RPD) has received different critisims, the most controversial being the choice of measure of performance, the Signal-To-Noise Ratio (SNR). The wrong choice this measure of performance, SNR, can lead to cross-talk (confusion) of factors affecting Sensitivity (Mean) and those affecting Dispersion. An example of right choice of SNR is given and also when SNR is wrongly chosen. The graphical method through gamma-plot gives the appropriate choice of SNR and consequently gives a clear cut among factors affecting Sensitivity and those affecting Dispersion.Item Comparison of the Invariant Solutions to the Multivariate Behren-Fisher Problems(Nigerian Association of Mathematical Physics, 2016) Oyeyemi, G. M.; Adebayo, P. O.Four of numerous solutions to multivariate Behrens-Fisher problem were selected for comparison. The four selected invariant solutions were; Yao, Johansen, James and Krishnamorrthy and Yu. Data were simulated to compare the four solutions under different distributions (Multivariate Beta, Multivariate Gamma and Multivariate Normal), sample sizes (N = 20, 30, 50, 100, 200, 400 and 600), number of variables (p = 2, 3 and 5) and for equal and unequal sample sizes. The comparisons were done at three levels of significance (alpha = 0.01, 0.025 and 0.05) using power of the test and type I error rate. The results showed that James procedure is better than all other procedures when p = 2 with small sample sizes, because it has second highest power with the lowest type I error rate. But when number of variables p = 3 or 4 and with large sample size, all the four procedures’ performances are the same.Item Comparisons of Some Outlier Detection Methods in Linear Regression Model(Faculty of Physical Sciences, University of Ilorin, Nigeria, 2017) Oyeyemi, G. M.; Oluwaseun, O. B.; Adeleke, M. O.Empirical evidence suggests unusual or outlying observations in data sets are much more prevalent than one might expect and therefore this paper addresses multiple outliers in linear regression model. Although reliable for a single or a few outliers, standard diagnostic techniques from an Ordinary Least Squares (OLS) fit can fail to identify multiple outliers. The parameter estimates, diagnostic quantities and model inferences from the contaminated data set can be significantly different from those obtained with the clean data. A regression outlier is an observation that has an unusual value of the dependent variable Y, conditional on its value of the independent variable X. Four procedures for detecting outliers in linear regression were compared; the Cook’s, DFFITS, DFBETAS, and Mahalanobi’s distances. DFBETAS is most efficient in outlier detection for small sample and small percentage of outliers but has low sensitivity when the sample size is large. Mahalanobi has more power of detection of small percentage of outliers regardless of sample size.Item Complex Survey Data Analysis: A Comparison of SAS, SPSS and STATA(Asian Network for Scientific Information, 2010) Oyeyemi, G. M.; Adewara, A. A.; Adeyemi, R. A.We compared three statistical packages (SAS, SPSS and STATA) in analzing complex survey data in the context of multiple regression analysis using concrete examples from two national healthcare database (MEPS and NDHS). The three packages are found to be efficient and flexible in analyzing complex survey data, but SAS in some cases seems to over estimate the variances of the sample statistics. Adjustment for stratification (incorporating stratification) is very important in complex survey analysis, especially if the stratification variable is endogenousItem Compliance with Road Safety Regulations among Commercial Motorcyclists in Nigeria(Canadian Academy of Oriental and Occidental Culture., 2012) Arosanyin, G. T.; Olowosulu, A. T.; Oyeyemi, G. M.Motorcycles account for one out of every four vehicles involved in crashes in Nigeria. The basic question has always been that do these motorcyclists comply with basic requisite safety rules? This paper therefore examined the level of compliance with some basic road traffic regulations among commercial motorcyclists commonly called Okada riders. The data for the analysis were collected from 334 commercial motorcyclists from Samaru-Zaria in Northern Nigeria through structured questionnaire triangulated with observation and inspection. The data were analysed using descriptive statistics and phi- coefficient. The study found total (100%) compliance with minimum age limit, number plate registration and motorcycle engine capacity but found 64, 16 and 45 per cent compliance rate with driver license, crash helmet usage and legal passenger permissive respectively. The phi coefficients reveal that there is no single factor that has all-through association with license holding, helmet usage and Highway Code awareness. Union membership has negative association with two of the cases, which confirms the antiregulation compliance posture of the okada union in the study area. The paper recommends the overhauling of the driver licensing system; enforcement of road traffic rules to enhance compliance and enactment of state traffic regulations to reinforce the national regulations.Item Computing Power and Sample Size for Hotelling T2 Test(Published by Bachudo Sciences Co. Ltd, 2007) Oyeyemi, G. M.Multiple comparisons often arises in statistical analysis and several methods exist for adjusting statistical significance levels taking into consideration the sample size and the power of the test. ln this paper Hotelling T-Square method for handling multiplicity in the p-variate two sample cases is discussed. A generalized method is derived for computing the exact power for the test. Apart from sample sizes, the magnitude and the direction of correlation between the variables also contributed to the power size of the test.Item A Conditional Restricted Equilibrium Correction Model on Nigerian Stock Exchange All-Share Index and Macroeconomic Indicators with 2008 Global Financial Crisis Effects: A Univariate Framework Approach(Scientific and Academic Publishing, 2015) Lawal, G. O.; Aweda, N. O.; Oyeyemi, G. M.This paper employed the modified autoregressive distributed lag (ADRL) procedure to establish a univariate single level relationship existing between the Nigerian Stock Exchange (NSE) All-Share Index and three macroeconomic indicators such as Treasury bill rate, nominal exchange rate and inflation rate in Nigeria. A conditional restricted equilibrium correction model (ECM) was postulated with significant long-run relationship between NSE All-Share Index, exchange rate and inflation rate. The model relates exchange rate and inflation rate negatively with the All-Share Index in the long-run. Treasury bill rate have no long-run relationship with All-Share Index. The short-run dynamics indicated a negative causal relationship between All-Share Index and the three macroeconomic indicators. The results of this paper showed that All-Share index is slow to react to any disequilibrium caused by shocks on these macroeconomic indicators in the long-run. The 2008 global financial crisis had an insignificant negative effect on the NSE All-Share Index due to improved financial deepening. Monetary policy stability is crucial to price level control because inflation is a monetary phenomenon in Nigeria. Therefore, this paper proposed that the efficient use of Treasury bills as apparatus of monetary policy (inflation-targeting) and major source of government financing is essential to the growth of the Nigerian stock market. In addition to efficient monetary policy through interest rate and most importantly exchange rate, a secure fiscal discipline through effective government spending will likely have a positive effect on the All-Share Index rapidly and directly.