Department of Mathematical Sciences

Permanent URI for this collection

https://repository.run.edu.ng/handle/123456789/73

Browse

Now showing 1 - 20 of 161

A Lotka-Volterra Non-linear Differential Equation Model for Evaluating Tick Parasitism in Canine Populations
(Mathematical Modelling of Engineering Problems, 2023-08-30) Adesina, Olumide Sunday
This research employs a modified version of the Lotka-Volterra non-linear first-order ordinary differential equations to model and analyze the parasitic impact of ticks on dogs. The analysis reveals that fluctuations in pesticide effects significantly influence tick populations and the size of the canine host. The study also uncovers that alterations in the size of the interacting species can lead to both stable and unstable states. Interestingly, in a pesticide-free environment, a decline in the inter-competition coefficient catalyzes an increase in the sizes of both interacting species. This increase, although marginal for the tick population, contributes to overall system stability. The findings underscore the utility of the Lotka-Volterra non-linear first-order ordinary differential equations in modeling the parasitic effect of ticks on dogs. To protect pets, particularly dogs, from the harmful effects of tick infestation, this study recommends the appropriate and regular application of disinfectants.
A modified generalized class of exponential ratio type estimators in ranked set sampling
(Scientific African, 2023-03-22) Adesina, Olumide Sunday
Background: Researchers consider ranked set sampling (RSS) as an alternative to simple random sampling (SRS) for data collection because studies have shown that it is more efficient and less biased. Also, introducing population parameters to estimators increases the efficiency of such estimators. Aim: This study derived a modified generalized class of exponential ratio estimator in RSS by introducing available population parameters and compared the results with an existing version in SRS. Methodology: The biases and mean square errors (MSE) of the proposed estimators were derived up to the terms of first-order approximation using Taylor’s series expansion. Efficiency was used as the mode of comparison between the proposed and existing estimators. Results: Life data sets and simulated data supported the numerical illustration to corroborate the theoretical results. Conclusion: The MSEs of the modified generalized class of estimators under RSS were found to be smaller than those of the existing generalized class of estimators under SRS; hence they are more efficient estimators.
A non-parametric analysis of the effect of Covid-19 pandemic on Nigerians’ well-being based on geopolitical zones
(JP Journal of Biostatistics, 2024-02-12) Adesina, Olumide Sunday
The COVID-19 pandemic has crippled the economic activities of so many nations across the globe since its outbreak in 2019. This study is focused on the consequential effect of the COVID-19 pandemic in terms of standard of living, and perception of economic and security situation of Nigerians. A non-parametric approach was adopted on a primary data obtained through the administration of questionnaires online by NoiPolls during the COVID-19 period. The result obtained from this study depicts that there is a significant relationship between the security situation and perception of the country’s economic situation. Kruskal-Wallis test was used to check if there is a significant difference in economic perception, security situation and standard of living where it was observed that there exists a significant difference based on the geopolitical zone. We, therefore, recommend that efforts should be made by the government towards improving the economic state of affairs especially in the southern part of the country as this will in the long run lead to sustainable cities and communities across the geopolitical zones which is one of the goals of SDGs. Moreover, efforts should also be made by the government towards improving the security situation in the north and southeast as improvement in the country’s economic situation has a direct influence on the security position of the country.
Adaptive Models for Tails of Distributions
(International Journal of Statistics and Economics, 2019) Adesina, Olumide Sunday
Adaptive Regression Model for Highly Skewed Count Data.
(IAEME, 2019-01-11) Adesina, Olumide Sunday
A big task often faced by practitioners is in deciding the appropriate model to adopt in fitting count datasets. This paper is aimed at investigating a suitable model for fitting highly skewed count datasets. Among other models, COM-Poisson regression model was proposed in this paper for fitting count data due to its varying normalizing constant. Some statistical models were investigated along with the proposed model; these include Poisson, Negative Binomial, Zero-Inflated, Zero-inflated Poisson and Quasi- Poisson models. A real life dataset relating to visits to Doctor within a given period was equally used to test the behavior of the underlying models. From the findings, it is recommended that COM-Poisson regression model should be adopted in fitting highly skewed count datasets irrespective of the type of dispersion.
Analyses of Spatial Variations of Kenaf in Experimental Field
(2010) Adekeye, Kayode
Preliminary investigations of experimental field usually involve collection of soil samples at widely spaced locations which are patchily or globally at variant spatially. This study was carried out to evaluate spatial variations in experimental fields using a split plot experiment distributed in a completely randomized design at Ikenne and Ilora between June and September 2006 (test crop was kenaf). The preliminary descriptive statistics suggested the dependency of the stem girth and height on the spatial positions. The variance - covariance analyses matrices of the plots showed that stem girth and plant height were independently distributed and exhibited a non stationarity principle. The results also revealed that spatial autocorrelation exists in patches in the experimental fields while the entire plots showed random distributions because the autocorrelatons were neither dominated by negative nor positive correlation and more than 50% of these values fall within the range of ± 2√n. From this study, a regionalized spatial variation is imminent in 625 m2 experimental plot despite the difference in the treatments. Spatial variations study was found necessary in any plot not more than an acre (250 m2) of land otherwise such variations should be treated as block or environmental effect(s).
Approximation techniques for maximizing likelihood function of generalized linear mixed models for binary response data
(sciencepubco, 2018-12-30) Adesina, Olumide Sunday
Evaluating Maximum likelihood estimates in Generalized Linear Mixed Models (GLMMs) has been a serious challenge due to some integral complexities encountered in maximizing its likelihood functions. It is computationally difficult to establish analytical solutions for the integrals. In view of this, approximation techniques would be needed. In this paper, various approximation techniques were exam-ined including Laplace approximation (LA), Penalized Quasi likelihood (PQL) and Adaptive Gauss-Hermite Quadrature (AGQ) tech-niques. The performances of these methods were evaluated through both simulated and real-life data in medicine. The simulation results showed that the Adaptive Gauss-Hermit Quadrature approach produced better estimates when compared with PQL and LA estimation techniques based on some model selection criteria.
Assessing the Impact of ATM and POS Transactions on Currency Circulation in Nigeria: a comparison of artificial neural network and linear regression models
(Revue d'Intelligence Artificielle, 2024-02) Adesina, Olumide Sunday
This work is aimed at assessing the impact of ATM and POS transactions on currency in circulation in Nigeria via the comparison of the Artificial Neural Network and Linear Regression models. The Pearson product moment correlation coefficient was used to assess the degree of relationship between currency in circulation and ATM transactions, currency in circulation and POS transactions, ATM transactions and POS transactions. In addition to this, a multiple linear regression model was also fitted on the data to in order to assess the impact of currency in circulation on ATM transactions and POS transactions. From the results obtained, it was observed that POS transactions have a significant impact on the amount of currency in circulation compared to ATM transactions because the value of POS transactions has a P-value of 1.5542E-13 which is significant at 0.05. This was further studied using the Artificial Neural Network (ANN). The Multilayer preceptor of ANN was adopted, with 70 percent of the data initially subjected to training and 30 percent for validity. Thereafter, the amount of training data was increased to 61.4% and testing data was increased to 38.6% and it was observed that there was corresponding increase in the R-square from 0.809 to 0.839. This shows that the R-square of the ANN can be improved by increasing the amount of training data and testing data.
Assessing the Role of Trade Liberalization in Facilitating Trade Flows and Economic Expansion: Evidence from ECOWAS Countries
(Springer, 2021-02-11) Adesina, Olumide Sunday
Assessment of noise-levels of generator-sets in seven cities of South-Southern Nigeria,
(Taylor & Francis, 2018-01-21) Adesina, Olumide Sunday
Noise pollution has been shown to be a global health hazard and this could be aggravated by the use of noise-emitting generators. Therefore, this study aims to determine the Sound Pressure Levels (SPLs), Sound Power Levels (LW) as well as Noisiness of sixty different models with various Power-ratings from fourteen generator brands, commonly used in homes/offices in seven cities covering South-Southern Nigeria. The results obtained between January 2013– December 2015 showed that for nearly all generator brands, models and ratings, the values of SPL were above the Permissible Noise Exposure Limits (PNELs) recommended by WHO, USEPA and EN of 90dB(A), 75dB(A) and 70dB(A) respectively for 8hour daytime safe human exposure. Also, the ‘Wilcoxon Signed Ranks Test’ analyses showed that the three (3) alternative hypotheses [Ha : SPL . 90dBA], [Ha : SPL . 75dBA] and [Ha : SPL . 70dBA] are statistically-significant and should be accepted – further implying that: the decibel-ratings of majority of these generator-models are evidently hazardous.
Assessment of Statiscal Analysis as a Tool for Interdisciplinary Research
(Journal of Science and Technology Research, 2018) Okewole, Dorcas Modupe
This study is focused on reviewing and analysing the frequency of use of statistics by researchers across different disciplines. The importance of statistics in planning, collection, presentation, analysis, interpretation and dissemination of data as useful and reliable information necessitates its involvement in research in all disciplines. This study serves to provide information on the current level of involvement of statistics in this regard and to sensitize non-users about it. A survey of two ISI and Scopus listed Journals for 2017 in each of various disciplines of the sciences, social sciences, engineering, humanities, education, medical sciences and agricultural sciences was carried out. The use or non-use of statistical applications was the focus of the survey. The chart drawn on the result showed that one of each of the journals in arts (7.1%), history (0%) and the two journals in Religious studies (13.6% and 10.7%), had very low statistics applications. All other disciplines (14 out of the 17 disciplines surveyed) apply statistics more frequently. The ANOVA from the beta regression indicated a highly significant (p < 0.0001) difference in statistics application across the disciplines. The conclusion was that researchers not familiar with the use of statistics should learn from literature and take advantage of the high impact of statistics for better output in their research.
Assessment Question Type in A Statistics Course for Non Majors: Analysis of Students' Preference
(International Association of Statistics Education, 2021) Okewole, Dorcas Modupe
This study involves the analysis of students’ preference on the Assessment Question Type (AQT) in an applied statistics course for non majors. Developing interest in a statistics course may be greatly influenced by performance in statistics courses taken earlier or expectations on the one being considered. It is important that assessment of the students’ performance in the course be through a valid evaluation tool which includes appropriate question type. Students’ AQT preference information was obtained using an instrument designed for the purpose and was shown to vary mostly between multiple choice and essay questions. Proportion of scores from multiple choice questions in the final examination was modeled with beta regression. Students’ level of understanding of the course (as judged by the total score in the examination) was shown to influence proportion of marks from the two AQT employed. Implications of the results were discussed.
The Bayesian Approach To Multi-equation Econometric Model Estimation
(Journal of Statistical and Econometric Methods, 2014) Okewole, Dorcas Modupe
Bayesian Dirichet Process Mixture Prior for Count Data
(IAEME, 2018-12-13) Adesina, Olumide Sunday
Bayesian Estimation of an Over-identified Multi-equation Model in the Presence of Multicollinearity
(Journal of the Nigerian Statistical Association, 2013) Okewole, Dorcas Modupe
Multi-equation systems have wide applications in modeling Economic issues. The Bayesian approach received very little attention in the past but is now gaining popularity with extensive application to areas hitherto handled by the classical method. The increasing interest is as a result of availability of numerical intensive software capable of solving intractable or complex numerical integration and other mathematical or computational difficulties. Violations of the assumptions underlying the models often arise in actual observed data. Multicollinearily is one of such violations which several researches have shown classical estimation approaches to he sensitive to. Studies on the performance of the Bayesian approach to such violations are however limited. This paper presents a Monte Carlo study of the Bayesian approach to multi-equation models estimation in the presence of multicollinearity. The mean, bias and MSE were used to compare the performance of the Bayesian approach to that of some classical approaches. A number of research scenarios were specified depicting presence and absence of multicollinearity. MSE from the scenario representing absence of multicollinearity was smaller than that from the scenario representing presence of multicollinearity. Results from the Bayesian approach in run 1 (representing presence of multicollinearity) showed that MSE for fir(one of the correlated exogenous variables) are 0.2825, 0.1128, 0.1079 and 0.0649 for sample sizes 20, 40, 60 and 100 respectively, whereas, they were 0.2503, 0.0642, 0.0406 and 0.0414 in the absence of multicollinearity represented by run 2. MSE for fl„from the classical approach were 0.4230, 0.1583, 0,1498 and 0.0897 for sample sizes 20, 40, 60 and 100 respectively, whereas, they were 0.3639, 0.0837, 0.0517 and 0.0540 in run 2. MSE from the Bayesian approach were smaller than those from the classical approach. The results showed that the Bayesian approach is less sensitive to multicollinearity in estimating the coefficients of exogenous variables of over-identified model.
A Bayesian Logistic Regression on Sickle Cell Anaemia: Variation in Survival by Age and Sex
(International Journal of Innovative Research & Development, 2017-11) Okewole, Dorcas Modupe
A study on variation in survival of sickle cell anaemia by age and sex was carried out. The dichotomous nature of the response variable (survival) suggested modelling with the logistic function. There is a dearth of Studies on the specific way in which survival of sickle cell anaemia varies across sex and different age groups of patients from the South-western part of Nigeria. Adequate information on the pattern of response of patients in the presence of significant covariates will yield more effective medical intervention. This study therefore was focused on the survival of sickle cell anaemia across sex and different age groups. The goal was to find out which of these two variables contribute significantly to the survival of sickle cell anaemia and the specific pattern across the different age groups. Data on Sickle cell anaemia patients obtained from the University College Hospital, University of Ibadan, in the South-western part of Nigeria was analysed in the Bayesian logistic Regression approach. Survival varied significantly across different ages, the estimate of the coefficient was -0.04215 with a Bayesian confidence interval (-0.04215, -0.02338). For the female, the percentage alive reduced at the child bearing ages compared with other ages. Sex on the other hand did not significantly contribute to the variation in survival, having the estimate of its coefficient as -0.1298 with a Bayesian confidence interval (-0.13 and 0.452). Age-specific medical attention might increase the life expectancy of these people.
Bayesian Multilevel Models for Count Data
(Nigerian Society of Physical Sciences, 2021-08-29) Adesina, Olumide Sunday
The traditional Poisson regression model for fitting count data is considered inadequate to fit over-or under-dispersed count data and new models have been developed to make up for such inadequacies inherent in the model. In this study, a Bayesian Multi-level model was proposed using the No-U-Turn Sampler (NUTS) sampler to sample from the posterior distribution. A simulation was carried out for both over-and under-dispersed data from discrete Weibull distribution. Pareto k diagnostics was implemented, and the result showed that under-dispersed and over-dispersed simulated data has all its k value to be less than 0.5, which indicates that all the observations are good. Also, all WAIC were the same as LOO-IC except for Poisson in the over-dispersed simulated data. Real-life data set from National Health Insurance Scheme (NHIS) was used for further analysis. Seven multi-level models were fitted and the Geometric model outperformed other models.
Bayesian Optimization for Parameter of Discrete Weibull Regression
(Journal of Advances in Mathematics and Computer Science, 2020-01) Okewole, Dorcas Modupe
This study aim at optimizing the parameter θ of Discrete Weibull (DW) regression obtained by maximizing the likelihood function. Also to examine the strength of three acquisition functions used in solving auxiliary optimization problem. The choice of Discrete Weibull regression model among other models for fitting count data is due to its robustness in fitting count data. Count data of hypertensive patients visits to the doctor was obtained at Medicare Clinics Ota, Nigeria, and was used for the analysis. First, parameter θ and β were obtained using Metropolis Hasting Monte Carlo Markov Chain (MCMC) algorithm. Then Bayesian optimization was used to optimize the parameter the likelihood function of DW regression, given β to examine what θ would be, and making the likelihood function of DW the objective function. Upper confidence bound (UCB), Expectation of Improvement (EI), and probability of Improvement (PI) were used as acquisition functions. Results showed that fitting Bayesian DW regression to the data, there is significant relationship between the response variable, β and the covariate. On implementing Bayesian optimization to obtain parameter new parameter θ of discrete Weibull regression using the known β, the results showed promising applicability of the technique to the model, and found that EI fits the data better relative to PI and UCB in terms of accuracy and speed.
Bayesian Optimization for Parameter of Discrete Weibull Regression
(SCIENCEDOMAIN International, 2020-03-13) Adesina, Olumide Sunday
This study aim at optimizing the parameter θ of Discrete Weibull (DW) regression obtained by maximizing the likelihood function. Also to examine the strength of three acquisition functions used in solving auxiliary optimization problem. The choice of Discrete Weibull regression model among other models for fitting count data is due to its robustness in fitting count data. Count data of hypertensive patients visits to the doctor was obtained at Medicare Clinics Ota, Nigeria, and was used for the analysis. First, parameter θ and β were obtained using Metropolis Hasting Monte Carlo Markov Chain (MCMC) algorithm. Then Bayesian optimization was used to optimize the parameter the likelihood function of DW regression, given β to examine what θ would be, and making the likelihood function of DW the objective function. Upper confidence bound (UCB), Expectation of Improvement (EI), and probability of Improvement (PI) were used as acquisition functions. Results showed that fitting Bayesian DW regression to the data, there is significant relationship between the response variable, β and the covariate. On implementing Bayesian optimization to obtain parameter new parameter θ of discrete Weibull regression using the known β, the results showed promising applicability of the technique to the model, and found that EI fits the data better relative to PI and UCB in terms of accuracy and speed.
Bayesian Regression Model for Counts in Scholarship
(IISTE, 2017-07-30) Adesina, Olumide Sunday
Discrete Weibul (DW) is considered to have the ability to capture under and over-dispersion simultaneously and also have a closed-form analytical expression of the quantiles of the conditional distribution. There is a need to further investigate how effective the model is, as compared to other competing models in the context of classical and Bayesian technique. In this study, the strength of DW is investigated, for both on frequentist and Bayesian technique. The Bayesian DW adopts parameterization, which makes both parameters of the discrete Weibull distribution to be dependent on the predictors. Bayesian Generalized linear mixed model is also implemented and is compared with the BDW, since Bayesian generalized linear mixed model is known to be robust in handling over-dispersion in count data. A simulation study and real life data was carried out for over and under-dispersed count data. The empirical analysis shows the superiority of Bayesian Generalized linear mixed model over Bayesian DW in the case of over-dispersed data as identified in the simulated data and real life data, but not for under-dispersed data as in the case of simulated study.

Browse

Browsing Department of Mathematical Sciences by Title

Results Per Page

Sort Options