Royal Statistical Society Nigeria Local Group Annual Conference Proceedings

LARGE-SCALE SAMPLE SURVEY DATA: AN ALTERNATIVE TO POPULATION DATA FOR SUSTAINABLE DEVELOPMENT IN THE 21ST CENTURY

Ahmed Audu — 2022-09-09

Socio-economic development has been defined as a progress in terms of economic and social factors within a geographic unit. Economic development implies raising the level of prosperity through increased production, distribution and consumption of goods and services. Social development, on the other hand, provides comparative information about citizens such as income, poverty, employment, employment security, education, health, crime and civic participation. Socio-economic development indicators have been conceptualized by several organizations like European Commission (2007), UNECE et al (2014) and OECD (2016) to include happiness, quality of life, well-being, living conditions, life situation, human development, social capital, generalized and political trust, environmental concepts (ecological footprints), concept of sustainable development etc

SIMILARITY ASSESSMENT FOR THE POPULATION DENSITY OF SOME CITIES IN NIGERIA USING DYNAMIC TIME WARPING ALGORITHMS

E. D Adedoyin — 2022-09-12

Similarity, or proximity, measures are used in diverse fields of inquiry to study the trajectory of random variables. Patterns, or trajectories, of human population density growth (measured with respect to constant area) are well-documented to be closely associated with social, economic and environmental development and vulnerabilities. This study is therefore aimed at investigating the trajectory of the population density of some Nigerian cities having population density ≥ 900 persons.km^-2 with a view of clustering the cities using Dynamic Time Warping (DTW) algorithms as the distance measure. The cities considered were selected on the basis of 2006 National Population Commission Census’ report. A Preliminary investigation for the optimal cluster number using K-means, Partition Around Medoids (pam) and Agglomerative Hierarchical algorithms showed that k = 2 and k = 6 produced optimal clusters. Since a higher optimal cluster value connote production of better grouping, outputs for k = 6 was selected. The result showed a dissimilar population density trajectory for Okene and Zaria while Uyo and Ikorodu cities had similar population density trajectory for the periods considered. Although Kano showed similar population density trajectory with Aba and Enugu, the cities of Enugu and Aba had more similar density trajectory than the city of Kano.

Valuation of Nigerian Crude Oil using Heston’s Stochastic-Jump Model

A. C. Onyegbuchulem — 2022-09-09

Presence of new information in many commodity markets leads to unpredictably changes and jumps in commodity prices. This article investigates the possible existence of jumps in the crude oil prices in Nigeria, an improved version of the Heston’s model was developed to take care of these changes and jumps. We present the model and demonstrate its implementation using crude oil price data from 2010 to 2020. Characteristic function approach was adopted to value oil prices. The Non-linear least squares method of calibration was used for the calibration of the model parameters. Simulation studies show that Heston Stochastic-Jump model is suitable for pricing crude oil in Nigeria. The Matlab software was used for both calculations and simulations

Statistical Study on the Impacts of Climatic Factors on the Prevalence of Malaria (A Case Study of Kaduna State)

Jamilu Yunusa Falgore — 2022-09-09

Despite immense efforts to rapidly stem the tide of malaria occurrences across the world, climatic factors and weather patterns continue to be major contributors to malaria transmission, especially within vulnerable communities of the world. The purpose of this research is to look into the statistical relationship between climatic variables and malaria presence in Kaduna State, Northwestern Nigeria. For ten years, from 2011 to 2020, data on malaria cases, temperature, and relative humidity were collected monthly in Kaduna state. To assess the relationship, level of association, and model fitness with the dataset, simple and multiple linear regression, correlation coefficients, and coefficient of determination were used. According to the findings, there was a significant correlation between humidity and malaria cases. This implies malaria cases were more prevalent, months humidity were high, as temperatures exhibit a fair negative relationship with malaria cases. As a result, we recommended that greater emphasis be placed on vector control activities and raising public awareness about the proper use of intervention strategies such as interior residual sprays to mitigate the epidemic, particularly during peak periods with favorable weather conditions.

STATISTICAL ANALYSIS OF PEOPLE LIVING WITH TUBERCULOSIS IN ONDO STATE

Oluwatoyin Kikelomo Bodunwa — 2022-09-09

This study reported cases of Tuberculosis diseases and those that co-infected with HIV among different age groups and gender between 2016 and 2020 in Ondo State, Nigeria. Chi-square test was used to investigate the relation that exist between the gender of TB patients and the HIV status. The result showed that the proportions of gender is found to be significant meaning that TB infection rate is higher in males compared to female. Also, 10% were co-infected with HIV and 90% were not. The difference in the proportions is found to be significant meaning that there were fewer number of patients that were co-infected with HIV as compared to those that were not. Also the proportions of those that were Pulmonary TB and Extra pulmonary was significant which depicts that there were fewer Extra pulmonary TB cases as compared to Pulmonary. This supports the fact that Pulmonary Tuberculosis which is contagious is expected to occur more frequently than extra pulmonary tuberculosis (non-contagious). The relationship between female TB patients that were likely to be co-infected with HIV than male TB patients was found to be significant.

A SCALED CONJUGATE GRADIENT METHOD USING THE DFP UPDATE FOR UNCONSTRAINED OPTIMIZATION PROBLEMS

I. A. OSINUGA — 2022-09-09

The Conjugate Gradient (CG) and Quasi-Newton methods are famous methods for finding solution to optimization problems involving large variables such as problems in optimal inflation rate, minimal cost, maximal profit, minimal error, optimal design and many more. In this presentation, we propose a modification of the hybrid Davidon-Fletcher-Powell-Conjugate-Gradient (DFP-CG)methods developed by Wan Osman et.al (2017) by adapting a spectral-scaling memory less DFP-update. The numerical implementation of the proposed method on a some selected unconstrained optimization test problems by adopting the performance profile by Dolan et al. (2002) indicates that the newly suggested method is competitive, robust and in most instances more efficient when compared with some existing CG methods in the literature.

Estimation of Mean in Single-phase Sampling with High and Low Extreme Maximum values using auxiliary information

Peter I. Ogunyinka — 2022-09-09

Significant improvement has been introduced to regression-in-ratio estimators in simple random sampling. However, such improvement will be jeopardized when there is extreme maximum or minimum value in survey data. This study has proposed three improved regression-in-ratio estimators that would correct the over-estimation or under-estimation effect as a result of extreme maximum or minimum values in survey data, respectively. The bias and the mean square error expressions were established for comparison of the proposed estimators. Theoretical comparison and empirical comparison, through simulation for twenty six populations with high and low extreme maximum values, confirmed that the proposed estimators were, generally, efficient over the reviewed estimators. Though, the proposed estimators were less bias to the reviewed estimators, but they were confirmed to be asymptotically efficient. Suggestion for further study in the detection of significant extreme values in sample survey data was proposed.

TRANSFER FUNCTION MODELLING OF INFLATION RATE AND IMPORT DUTIES IN NIGERIA

Wiri Leneenadogo — 2022-09-09

The goal of this investigation is to use transfer function to model Nigeria's inflation rate and import duties. The two series were collected from the website of Nigeria's Central Bank (for 16 years). The data were checked for stationarity using appropriate transformations at first. The ARIMA (p d q), was used to estimate five models for the input series. The best model was selected based on the minimum information criterion. As a result, the input series was modelled using ARIMA (1,1,0) autoregressive integrated moving average models. The input and output variable was pre whitened. The analysis of cross-correlation function was used to provide a rational for the polynomial representation of the dynamic transfer function models. It was discovered that the calculated noise was autocorrelated. This filled in the gaps in the transfer function model, which was then used to fit the whole thing together. The resulting model was put through a diagnostic test and found to be suitable. As a result, a forecast was made.

RESTRICTED MULTIVARIATE-GARCH) MODEL OF EXCHANGE RATE, INFLATION RATE AND CRUDE OIL PRICE IN NIGERIA

Leneenadogo WIRI — 2022-09-09

In this study, the micro- economics in Nigeria were modelled using restricted multivariate Generalized Autoregressive conditional heteroscedasticity (M-GARCH). This strategy was chosen because the restricted technique is smoother than the unconstrained approach. The original series' time plot revealed trend, and the return series' logarithm transformation revealed stationarity. The diagonal VECH and BEKK models were estimate using the return series. However, for both Diagonal VECH and Diagonal BEKK, three conditional variances (GARCH) models and three co-variance models were computed. All of the variance and covariance models were significant at 5%. The Diagonal VECH model's parameters were all significant, although not all of them were positive definite. BEEK MODELS are required because the parameters of the square matrix of the Diagonal VECH model are not positive definite. All parameters in the BEKK model of the currency rate, inflation rate, and crude oil prices are significant and positive definite. Finally, the Diagonal BEKK model is the appropriate model for estimating the variance and co-variances of the exchange rate, inflation rate, and crude oil prices in Nigeria.

STATISTICAL MODELLING OF GENE REGULATORY NETWORKS

Omolola Dorcas Atanda — 2022-09-09

Biological network analysis is a rapidly growing field which is increasing our understanding of biological process. The study and modeling of biological networks are important in life science today. A wide range of scientists are interested in quantifying the link between nodes in a system, however the linkage is not as straight forward as it might seem. The challenge is, how to extract relevant information and translate this information to knowledge that can yield clinically actionable results. Insights gained from successful computational statistics of networks topology can, in principle, be used to design new experiments that test these insights in a broad context. In this work, a newly derived discretised Power probability density function is proposed for in-degree distribution of gene regulatory networks. A statistical comparison of the newly proposed degree distribution was made with alternative degree distribution in literature.

Comparison of different learning rate (step size) on Logistic regression using FR conjugate gradient optimizer

Ibidapo D Dada — 2022-09-09

Conjugate gradient algorithm is one of the effective optimization algorithms used in solving logistic regression problems. This paper is focused on comparing some existing learning rate methods to reduce the objective function value of the logistic regression model with a limited number of iterations and reduced processing time. Fletcher-Reeves (FR) conjugate gradient method was run in python program using admission and iris flowers dataset to examine the performance of each learning rate. The numerical results of each step size were compared. The result shows that Armijo step size performs better in terms of number of iterations and processing time with good model accuracy.

PATH ANALYSIS OF FACTORS INFLUENCING DEPRESSION AMONG FUNAAB UNIVERSITY STUDENTS

O. A. Wale-Orojo, — 2022-09-09

Depression has become a prevalent mental illness that affects one out of every twenty persons at one point in their lifetime. It has been reported that depression can develop at any age and there are quite a number of factors that could lead an individual into depression. The aim of this research is to establish the effect of four selected factors influencing depression among FUNAAB students namely; Sexual Risk Behaviour-, Alcohol usage-, Bullying-Victimization- and Academic Performance -. Studies have examined the effects of each factor on depression separately, but this study improved on these by investigating the total, direct and indirect effect of each and all these factors on Depression using Path Analysis. A well-structured questionnaire was developed to source for data. The instrument reliability was checked for effectiveness using Cronbach Reliability Test. 39.6% of the respondents’ showed signs of depression; Academic Performance, Alcohol usage and Bullying-victimization all contributed significantly to the presence of depression. While a unit increase in others will increase the level of depression, a good academic performance will reduce the effect and chances that a student will be depressed. Hence, efforts should be made to help cushion the effect of the independent factors on depression.

A Cauchy Transformation approach to the Robustness of Quantile Regression Model to Outliers

Joy C. Nwabueze — 2022-09-09

The main advantage of quantile regression models had over OLS is their robustness to outliers. This is because quantile regression models are insensitive to outliers and skewed distributions. This very property of quantile regression model is the same with the sample quantile. This work was done to examine the robustness of quantile regression model to outliers. Data analysis was done usingreal life data set on fuel consumption (in miles per gallon), in highway driving as the response variable. Extreme values where inserted to create outliers in the response variable data set. Car weight, length, wheel base, width, Engine size and horse power are the explanatory variables used in the analysis with a sample size of 91. The standard Cauchy distribution was used to transform the quantile regression model. The results show that the graphs of the mean square errors clustered around the zero line in all the study quantiles, also the descriptive results show that the residual means is equal to the residual medians and equal to zero. The skewness of the residuals approximates to zero across all the study quantiles, while the kurtosis approximates to 3, both the residual standard deviations, mean square errors and root mean square errors approximate to zero across all the study quantiles. From the results of the analysis, it can be concluded that quantile regression model is insensitive to outliers.

A MODIFIED EXPONENTIAL-TYPE ESTIMATOR FOR POPULATION MEAN WITH TWO AUXILIARY VARIABLES IN TWO PHASE SAMPLING

O. M. Olayiwola — 2022-09-09

The ratio estimators of finite population mean are applicable in various fields such as health, education, agriculture etc. Because one of the parameters used by this field in decision making includes the mean and very efficient estimation of this parameter will improve the outcome of such decision. Ratio estimators are applicable when the population mean of auxiliary variable X is known. But in real life situation complete information about population mean are usually unavailable which make ratio estimator impracticable. However, the ratio estimator is always bias and sometimes less efficient. In other to improve the efficiency and reduce the biasedness of existing estimator, the proposed estimators were modified using exponential as one of the improvement strategies. The inability to apply estimators in real-life scenarios stems from a lack of understanding of the auxiliary variable's population mean. The effectiveness of a class of a finite population mean in a double sampling estimator is investigated when the population mean of an auxiliary variable is estimated using exponential-type based on a preliminary large sample. Taylor series expansion up to second degree approximation was used to obtain the suggested estimators' biases and mean square errors. The suggested estimators' efficiency was compared to that of some relevant current estimators in an empirical study using four (4) real life data sets. The suggested estimators have lower mean square errors and higher percentage relative efficiencies than related estimators investigated in the study, according to the results. In addition, the present research can be used in any area of estimation and study variable. The estimators modified in this research work can be used in sampling theory at the estimation stage.

STATISTICAL ANALYSIS OF FACTORS AFFECTING THE STANDARD OF LIVING IN NIGERIA

J. Y. Falgore — 2022-09-09

Despite the effort of the government to reduce or eradicate unemployment, they still lack strategies on how to generate employment. Hence, there is a need to conduct research on these factors that affect the standard of living and to find out between those factors which one have more impact on Nigeria’s standard of living. The method of data analysis adopted in this study is multiple linear regression with poverty as the dependent variable Y, inflation and unemployment are the two independent variables. The analysis of variance (ANOVA) techniques was employed to examine the significant difference between the response and the explanatory variables. The data used for this research were gotten from the World Bank database which covers the period of 45 years (1976-2020). The coefficients of the regression are significant with a P-value of 0.0001. We observe that the value of the correlation coefficient is 0.837. This indicates that there is a strong positive association between poverty and the unemployment rate, with a significant p-value of 0.001<0.05. Based on these results, we concluded that for Nigerians to live above poverty and for the economy to grow, the government must take concrete steps in opening up the economy.

STOCHASTIC MODELLING OF NIGERIA INFLATION RATE

A. A. Akintunde — 2022-09-12

Inflation rate is an essential economic variable. It gives a picture of the performance of an economy. Nigeria as a developing country needs to monitor the inflation rate with the other economic factors in order to ensure that the economy does not go out of control. The impact of the Corona Virus Disease 2019 (COVID-19) on the developing countries can be checked through such monitoring. This study examined the long run prediction of the dynamics of the Nigeria inflation rate using the Markov Modeling approach. The Nigeria inflationary system was considered as a three state Markov process and the system’s distribution was determined at the steady state. Result shows that the long run probability of increased inflation rate is high. The study recommends the use of Markov approach for modelling the behaviour of dynamic systems and that reversion mechanism be put in place especially by the governments to mitigate inflation rate in Nigeria.

COMPARISON AND EVALUATION OF DIFFERENT POST-HOC TEST STATISTIC IN ENGINEERING AND EDUCATION USING RANDOMIZED COMPLETE BLOCK DESIGN UNDER ASSUMPTION OF EQUAL VARIANCE.

S. Bukar — 2022-09-12

The high percentage of error rate generated by post-hoc test is one of the major challenge of researchers across different fields of enquiry as well as the correct method to adopt, this has lead to under utilization of the numerous methods, this research, aimed at evaluating and comparing various methods to determine the most robust, convenient, optimal and most efficient in detecting the least percentage comparison wise and experimental wise error rates under the condition of homogeneity of variance, The research used secondary data from the field of Engineering and Education all in the form of Randomized Complete Block Design, the data were subjected to homogeneity of variance test and the result confirm that their variances were equal (α >0.05),On the same vein the result was able to find out that violation of the assumption for homogeneity of variance has no any significant effect on the comparison wise and experimental wise error rates,Similarly the EER generated by the various methods were all less than the alpha level of 10% (<0.10%) ,this shows that all the methods are best when the researcher is interested in EER, While for the CER, all the methods kept the error rate a little above ten percent (>10%), except for Tukey and SNK. Therefore, these methods can be adopted in the fields of Engineering and Education depending on the researcher’s interest base on the error type, consequently when the researchers interest is to minimized both comparison and experimental wise error rate, the best methods across the research field is to adopt Tukey and SNK method as they kept the error rate below the chosen alpha level

PREFERENCE OF BAYESIAN METHODS OVER CLASSICAL METHOD IN ESTIMATING THE SCALE PARAMETER OF INVERSE RAYLEIGH FRECHET DISTRIBUTION

Akeem Ajibola Adepoju — 2022-09-13

Among the methods of parameter estimation, maximum likelihood approach is the most often used. However, Maximum likelihood function (MLE) is data dependence, and insufficiecncy of data may cause the results obtained from this method to be unreliable. In this case, the Bayesian method, which allows the usage of the prior knowledge on the parameters in the estimation process, is adopted. This research paper aims to study the Bayesian analysis and
compared it with maximum likelihood estimator on the scale parameter estimation of Inverse Rayleigh Frechet distribution based on uniform and quasi priors and applied Mean square error MSE criteria as a basis for comparison. In the Bayesian method, the Bayes estimates were obtained under Squared Error Loss Function (SELF), Quadratic Loss Function (QLF) and the Precautionary Loss Function (PLF). The performances of these estimators were compared to
the Maximum Likelihood Estimates based on simulation study. The results of this analytic simulation show that the quadratic loss function is the preferred estimators since its posseses the lowest mean Square Error (MSE) under uniform prior and quasi prior. Finally, the results show that quadratic loss functions under Uniform prior and Quasi prior outperformed the squared error loss function, the precautionary loss function and Maximum Likelihood estimator
across different sample sizes.

FLEXIBILITY OF GENERALIZED DISTRIBUTIONS

Ifeyinwa Vivian Omekam — 2022-09-12

Some existing distributions are limited in shapes of Probability Density Function (PDF) and Hazard Function (HF) which constrains their use in analysis of certain types of data. Generalizing these distributions often deal with this constraints on usage by introducing flexibility. Generalized distributions were derived using the Generalized Pareto Distribution (GPD) as base distribution. Exponentiated GPDs called Lehmann Type II GPD (LIIGPD) and Lehmann Type I GPD (LIGPD) having an additional parameter each were obtained by applying Lehmann Alternative 1 (LA1) and Lehmann Alternative 2 (LA2) parameter induction methods respectively. Flexibility of generalized distributions was established by comparing the shapes of probability density and hazard functions of LIIGPD and LIGPD with those of the GPD. No new probability density or hazard shape was introduced by LIIGPD but the new shape introduced by LIGPD demonstrated flexibility of generalized distributions. Generalized distributions do not always introduce new density and hazard shapes but often improve flexibility of distributions.

Parameter (shape) Estimation of Weibull-Exponential Distribution Using Classical and Bayansian Approach Under Different Loss Functions

A. A. Adepoju — 2022-09-12

The Bayesian as a statistical approach is a method applied in statistical inference which helps researchers to incorporate prior information surrounding the population parameter with support from information embodied in a sample to guide the inference process. From the Bayesian viewpoint, the choice of prior depends on the one’s wide knowledge of the subject matter, since there is no obvious approach from which one can decisively conclude that one prior has edge over the other. This paper aim at studied the parameter (shape) of Weibull-exponential distribution via classical and the Bayesian approach. Different estimates of the parameter (shape) were obtained from the Bayesian approach using quasi and extended Jeffery priors, under various loss functions. The results shows that the quadratic loss functions under extended Jeffrey prior and quasi prior outperformed the squared error loss function and the precautionary loss function across different sample sizes. The result also reveals that the Bayesian estimate of the parameter (shape) under extended Jeffrey and quasi prior using quadratic loss function is better than the maximum likelihood estimate. Finally, it was deduced that, an increment in the sample size, makes the error to reduce and the estimates approach the real value of the parameter (shape).

BRIDGING DEMOGRAPHIC DATA GAPS THROUGH POPULATION PROJECTIONS

Samuel Irobinso Kalu — 2022-09-12

The 2030 Agenda for Sustainable Development launched in 2015 aims to eradicate poverty and hunger, improve health and well-being, quality education, achieve greater gender equality among other goals in order to achieve sustainable development by 2030. Regrettably, the modest progress recorded prior to and since 2015 towards realizing the Sustainable Development Goals (SDGs) is being eroded by the covid-19 pandemic since 2020. Moreover, the adverse economic impact of the pandemic is likely to undermine the supply of funds for timely collection of good quality data required to guide policies and development planning. Using the 2006 Nigeria Census and the 2018 Nigeria Health and Demographic Survey data, this paper provides reliable population estimates and projections for Nigeria for the period 2015-2035. The data are also disaggregated into special population groups to provide requisite denominators to support calculation of appropriate indicators for effective tracking of progress towards achievement of the SDGs.

Factors Affecting Uptake of Optimal Doses of Tetanus Toxoid Vaccine in Six States of Nigeria

M. Hamaza — 2022-11-09

Background: Tetanus toxoid (TT) immunization for pregnant women has remained the most effective strategy in eliminating neonatal and maternal tetanus. The vaccine for the pregnant women is supposed to be given five times. Despite the availability of vaccines for almost 50years, tetanus in pregnant women still remained a public health problem and contributor to morbidity and mortality in Nigeria. Many qualitative and quantitative factors contribute to the continued existence of tetanus in pregnant women. The objective of this analysis is to identify individual characteristics that could be associated with maternal immunization against tetanus.

Methods: The data comes from the study conducted by the Federal Ministry of Health Nigeria in collaboration with the National Bureau of Statistics (NBS) and the World Bank in 2013. A total of 1504 pregnant women responded for tetanus related questions in the interview at ANC units in six states of Nigeria. The analysis involved the use of Kruskal Wallis analysis of variance to test the statistical significance difference between each pair of variables for the quantitative data, Chi – square to test the significant relationship between each pair of variables for the qualitative data, and Multinomial logistic regression for multivariate analysis.

Results: A total of 1504 pregnant women responded for tetanus related questions in the interview at ANC units in six states of Nigeria. Among the respondents, 42.1% of the women were from Adamawa state, 1.8% from Benue, 12.2% from Nassarawa, 6.8% from Ogun state, 25.9% from 0ndo and 11.2% from Taraba state. Almost 95.9% of the women were married with median age of 25years, ranging from 13 - 50years. About 39.3% of the women had secondary and higher education, 21.4% did not receive tetanus toxoid vaccine while only 21.4% received more than 2 times. Several factors such as use of tetanus toxoid injection, use of traditional birth attendant services, spouse education, primigravidae, and age, were found to be significant predictors of maternal immunization against tetanus.

Conclusion: The results shows the urgent need for the government to strengthen the immunization coverage in Nigeria, through enlightenment, providing more easily accessible centers, and providing more well trained personnel to carry out the exercise.