Oluchukwu Asogwa | University of Nigeria, Nsukka (original) (raw)

Uploads

Papers by Oluchukwu Asogwa

Research paper thumbnail of Categorical Analysis of Variance on Knowledge, Compliance and Impact of Hand Hygiene among Healthcare Professionals during COVID-19 Outbreak in South-East, Nigeria

Asian research journal of mathematics, Feb 21, 2023

Research paper thumbnail of Robustness of randomisation tests as alternative analysis methods for repeated measures design

Statistics in Transition New Series

Randomisation tests (R-tests) are regularly proposed as an alternative method of hypothesis testi... more Randomisation tests (R-tests) are regularly proposed as an alternative method of hypothesis testing when assumptions of classical statistical methods are violated in data analysis. In this paper, the robustness in terms of the type-I-error and the power of the R-test were evaluated and compared with that of the F-test in the analysis of a single factor repeated measures design. The study took into account normal and non-normal data (skewed: exponential, lognormal, Chi-squared, and Weibull distributions), the presence and lack of outliers, and a situation in which the sphericity assumption was met or not under varied sample sizes and number of treatments. The Monte Carlo approach was used in the simulation study. The results showed that when the data were normal, the R-test was approximately as sensitive and robust as the F-test, while being more sensitive than the F-test when data had skewed distributions. The R-test was more sensitive and robust than the F-test in the presence of a...

Research paper thumbnail of Prediction variance capability of orthogonal uniform composite designs and orthogonal array composite designs in the spherical region

Communications in Statistics - Theory and Methods

Research paper thumbnail of Of Students Academic Performance Rates Using Artificial Neural Networks (ANNs)

A model based on the multilayer perception algorithm was programmed. The result from the test dat... more A model based on the multilayer perception algorithm was programmed. The result from the test data evaluation showed that the programmed Artificial Neural Network model was able to correctly predict and classify the performance of students with Mean Correct Classification Rate CCR of 97.07%.

Research paper thumbnail of On The Comparison of Methods of Estimating Variance Components : A Case of Gudali Beef Cattle

31TP 2 PUqackasoo@yahoo.comU31T Abstract A two-stage nested design with unequal replications was ... more 31TP 2 PUqackasoo@yahoo.comU31T Abstract A two-stage nested design with unequal replications was used in this study with an intention to capture the variability effects in the model. Moreover, five different methods of variance component estimation which were randomly chosen from the frequency approach were compared in this study. These methods which were Analysis of Variance method (ANOVA), Quasi-maximum-likelihood method (QML), Modified likelihood method (ML), Restricted maximum-likelihood method (REML), and Modified maximum likelihood method (MML), recommended “modified maximum likelihood method as the best since it has the smallest minimum variances”.

Research paper thumbnail of On The Application of Queuing Model in Nigeria Banking Sector: a Case Study of Zenith Bank Plc, Abakaliki Branch

This study focused on the queuing system in Nigerian banking sector. The sources of data were pri... more This study focused on the queuing system in Nigerian banking sector. The sources of data were primary and were collected from Zenith Bank Plc, Abakaliki Branch at the withdrawal section of the banking hall for four days in a period of one month interval. The data were collected based on the arrival pattern and the service pattern of customers. The methodology employed followed by the birth and death Markovian process. The results obtained showed that the arrival rate is 0.2219 and the service rate is 0.1524. However, the probability that the servers are idle is 0.1574 and the cost incurred from waiting is N656.76. Besides, any increment on the number of servers by the bank management will help reduce the time customers spend on queue and also reduce cost incurred from waiting. It was also suggested that the management of the concerned Bank should employ only capable servers as to reduce customers wait on service, that is when the cause of long time taken to serve one customer is not...

Research paper thumbnail of Principal Component Factor Analysis of Some Development Factors in Southern Nigeria and Its Extension to Regression Analysis

This study was conducted to evaluate some development factors in Southern Nigeria in order to asc... more This study was conducted to evaluate some development factors in Southern Nigeria in order to ascertain common factors that explained the interrelationships among them and identify best cities for recommendation. A total sample of 250 cities from different states in three geopolitical zones in Southern Nigeria was used in this study and 11 development factors were considered. Kaiser-Meyer-Olkin (KMO) of (> 0.5) was computed to test the sampling adequacy; Bartlett’s Test of Sphericity (Significant at 0.001) was conducted to test whether the correlation between the variables are sufficiently large for factor analysis; correlation matrix was computed to confirm the inter-item correlation. In this analysis, principal component factor analysis was the factor extraction method. Varimax rotation technique was used for factor rotation. The result showed that three new factors with eigenvalues greater than 1 were successfully constructed. The three new factors accounted for 71.63% of tota...

Research paper thumbnail of On the Modeling of the Effects of COVID-19 Outbreak on the Welfare of Nigerian Citizens, Using Network Model

American Journal of Applied Mathematics and Statistics, 2020

A multilayer perception algorithm based model was established in this research. The result from t... more A multilayer perception algorithm based model was established in this research. The result from the test data evaluation showed that the established Artificial Neural Network model was able to correctly predict and classify the effects of COVID-19 outbreak on the well-being of Nigerian citizens with Mean Correct Classification Rate () of 98.05%. The value of the of the model which was classified as good (88.23%), also aligned with the result obtained from the Mean Correct classification rate. The architecture model also indicated a very high sensitivity and a low specificity values respectively. The study was able to show that some factors like economy, farming activities, religious activities, education, etc. were negatively affected whereas crime rates, unwanted pregnancy, relationship between parents and their children were positively influenced during the lockdown period of coronavirus pandemic outbreak in Nigeria.

Research paper thumbnail of On the Fourier Residual Modification of Arima Models in Modeling Malaria Incidence Rates among Pregnant Women

American Journal of Theoretical and Applied Statistics

Research paper thumbnail of On The Comparison of Methods of Estimating Variance Components: A Case of Gudali Beef Cattle

A two-stage nested design with unequal replications was used in this study with an intention to c... more A two-stage nested design with unequal replications was used in this study with an intention to capture the variability effects in the model. Moreover, five different methods of variance component estimation which were randomly chosen from the frequency approach were compared in this study. These methods which were Analysis of Variance method (ANOVA), Quasi-maximum-likelihood method (QML), Modified likelihood method (ML), Restricted maximum-likelihood method (REML), and Modified maximum likelihood method (MML), recommended " modified maximum likelihood method as the best since it has the smallest minimum variances ". Keywords: Two-way nested design, Analysis of variance method (ANOVA), Quasi-maximum-likelihood method (QML), Modified likelihood method (ML), Restricted maximum-likelihood method (REML), Modified maximum likelihood method (MML). 1.0 Introduction Nested design, otherwise called hierarchical design, is a form of experimental design with a prominent characteristic such that each level of every factor occurs with all levels of the other factors, without interactions. For instance, in certain types of studies, the levels of one factor B, will not be identical across all levels of factor A, each level of factor A will contain different levels of factor B, therefore, levels of B are said to be nested within the levels of factor A. However, the basic principle for estimating variance components has been, and to a large extent still is, that of equating quadratic functions of the observations to their expected values. Obvious candidates for such functions are those of the analysis of variance table. The first papers describing this procedure appear to be those of [7], whose interest was in weights of slubbings from the carding process in the woolen industry, and [28] who analyzed catches of different species in successive hauls of plankton nets. Although [7] and [28] represent both sides of the Atlantic, it appears that major developments in variance components estimation subsequently took place mainly in the U. S. A. An exception to this was [11], dealing with nested classifications, and then came [6] concerned with randomized complete blocks, and [24] dealing with approximate sampling distributions of estimated variance components. This was followed by [9] who put a firm foundation to the distinction between the fixed effects model (Model I) and the random effects model (Model II), a distinction which [29] later took great exception to.

Research paper thumbnail of 4-On-the-Efficiency-of-Some.pdf

This work examines the Relative Efficiency ( ) of some selected Randomized Complete Block Designs... more This work examines the Relative Efficiency ( ) of some selected Randomized Complete Block Designs ( ).The efficiency of the selected designs showed that design was the most preferable having Mean Square Error of 6.14, followed by design with Mean Square Error of 9.11 and design with Mean Square Error of 18.08. The results from the pairwise relative efficiency of the selected designs show that ( , ) = 0.34 with the smallest relative efficiency value and ( , ) = 2.94 with the largest relative efficiency value. We recommended design as the best design for this particular problem since its mean square error remains the smallest.

Research paper thumbnail of Of Students Academic Performance Rates Using Artificial Neural Networks (ANNs

A model based on the multilayer perception algorithm was programmed. The result from the test dat... more A model based on the multilayer perception algorithm was programmed. The result from the test data evaluation showed that the programmed Artificial Neural Network model was able to correctly predict and classify the performance of students with Mean Correct Classification Rate CCR of 97.07%.

Research paper thumbnail of Categorical Analysis of Variance on Knowledge, Compliance and Impact of Hand Hygiene among Healthcare Professionals during COVID-19 Outbreak in South-East, Nigeria

Asian research journal of mathematics, Feb 21, 2023

Research paper thumbnail of Robustness of randomisation tests as alternative analysis methods for repeated measures design

Statistics in Transition New Series

Randomisation tests (R-tests) are regularly proposed as an alternative method of hypothesis testi... more Randomisation tests (R-tests) are regularly proposed as an alternative method of hypothesis testing when assumptions of classical statistical methods are violated in data analysis. In this paper, the robustness in terms of the type-I-error and the power of the R-test were evaluated and compared with that of the F-test in the analysis of a single factor repeated measures design. The study took into account normal and non-normal data (skewed: exponential, lognormal, Chi-squared, and Weibull distributions), the presence and lack of outliers, and a situation in which the sphericity assumption was met or not under varied sample sizes and number of treatments. The Monte Carlo approach was used in the simulation study. The results showed that when the data were normal, the R-test was approximately as sensitive and robust as the F-test, while being more sensitive than the F-test when data had skewed distributions. The R-test was more sensitive and robust than the F-test in the presence of a...

Research paper thumbnail of Prediction variance capability of orthogonal uniform composite designs and orthogonal array composite designs in the spherical region

Communications in Statistics - Theory and Methods

Research paper thumbnail of Of Students Academic Performance Rates Using Artificial Neural Networks (ANNs)

A model based on the multilayer perception algorithm was programmed. The result from the test dat... more A model based on the multilayer perception algorithm was programmed. The result from the test data evaluation showed that the programmed Artificial Neural Network model was able to correctly predict and classify the performance of students with Mean Correct Classification Rate CCR of 97.07%.

Research paper thumbnail of On The Comparison of Methods of Estimating Variance Components : A Case of Gudali Beef Cattle

31TP 2 PUqackasoo@yahoo.comU31T Abstract A two-stage nested design with unequal replications was ... more 31TP 2 PUqackasoo@yahoo.comU31T Abstract A two-stage nested design with unequal replications was used in this study with an intention to capture the variability effects in the model. Moreover, five different methods of variance component estimation which were randomly chosen from the frequency approach were compared in this study. These methods which were Analysis of Variance method (ANOVA), Quasi-maximum-likelihood method (QML), Modified likelihood method (ML), Restricted maximum-likelihood method (REML), and Modified maximum likelihood method (MML), recommended “modified maximum likelihood method as the best since it has the smallest minimum variances”.

Research paper thumbnail of On The Application of Queuing Model in Nigeria Banking Sector: a Case Study of Zenith Bank Plc, Abakaliki Branch

This study focused on the queuing system in Nigerian banking sector. The sources of data were pri... more This study focused on the queuing system in Nigerian banking sector. The sources of data were primary and were collected from Zenith Bank Plc, Abakaliki Branch at the withdrawal section of the banking hall for four days in a period of one month interval. The data were collected based on the arrival pattern and the service pattern of customers. The methodology employed followed by the birth and death Markovian process. The results obtained showed that the arrival rate is 0.2219 and the service rate is 0.1524. However, the probability that the servers are idle is 0.1574 and the cost incurred from waiting is N656.76. Besides, any increment on the number of servers by the bank management will help reduce the time customers spend on queue and also reduce cost incurred from waiting. It was also suggested that the management of the concerned Bank should employ only capable servers as to reduce customers wait on service, that is when the cause of long time taken to serve one customer is not...

Research paper thumbnail of Principal Component Factor Analysis of Some Development Factors in Southern Nigeria and Its Extension to Regression Analysis

This study was conducted to evaluate some development factors in Southern Nigeria in order to asc... more This study was conducted to evaluate some development factors in Southern Nigeria in order to ascertain common factors that explained the interrelationships among them and identify best cities for recommendation. A total sample of 250 cities from different states in three geopolitical zones in Southern Nigeria was used in this study and 11 development factors were considered. Kaiser-Meyer-Olkin (KMO) of (> 0.5) was computed to test the sampling adequacy; Bartlett’s Test of Sphericity (Significant at 0.001) was conducted to test whether the correlation between the variables are sufficiently large for factor analysis; correlation matrix was computed to confirm the inter-item correlation. In this analysis, principal component factor analysis was the factor extraction method. Varimax rotation technique was used for factor rotation. The result showed that three new factors with eigenvalues greater than 1 were successfully constructed. The three new factors accounted for 71.63% of tota...

Research paper thumbnail of On the Modeling of the Effects of COVID-19 Outbreak on the Welfare of Nigerian Citizens, Using Network Model

American Journal of Applied Mathematics and Statistics, 2020

A multilayer perception algorithm based model was established in this research. The result from t... more A multilayer perception algorithm based model was established in this research. The result from the test data evaluation showed that the established Artificial Neural Network model was able to correctly predict and classify the effects of COVID-19 outbreak on the well-being of Nigerian citizens with Mean Correct Classification Rate () of 98.05%. The value of the of the model which was classified as good (88.23%), also aligned with the result obtained from the Mean Correct classification rate. The architecture model also indicated a very high sensitivity and a low specificity values respectively. The study was able to show that some factors like economy, farming activities, religious activities, education, etc. were negatively affected whereas crime rates, unwanted pregnancy, relationship between parents and their children were positively influenced during the lockdown period of coronavirus pandemic outbreak in Nigeria.

Research paper thumbnail of On the Fourier Residual Modification of Arima Models in Modeling Malaria Incidence Rates among Pregnant Women

American Journal of Theoretical and Applied Statistics

Research paper thumbnail of On The Comparison of Methods of Estimating Variance Components: A Case of Gudali Beef Cattle

A two-stage nested design with unequal replications was used in this study with an intention to c... more A two-stage nested design with unequal replications was used in this study with an intention to capture the variability effects in the model. Moreover, five different methods of variance component estimation which were randomly chosen from the frequency approach were compared in this study. These methods which were Analysis of Variance method (ANOVA), Quasi-maximum-likelihood method (QML), Modified likelihood method (ML), Restricted maximum-likelihood method (REML), and Modified maximum likelihood method (MML), recommended " modified maximum likelihood method as the best since it has the smallest minimum variances ". Keywords: Two-way nested design, Analysis of variance method (ANOVA), Quasi-maximum-likelihood method (QML), Modified likelihood method (ML), Restricted maximum-likelihood method (REML), Modified maximum likelihood method (MML). 1.0 Introduction Nested design, otherwise called hierarchical design, is a form of experimental design with a prominent characteristic such that each level of every factor occurs with all levels of the other factors, without interactions. For instance, in certain types of studies, the levels of one factor B, will not be identical across all levels of factor A, each level of factor A will contain different levels of factor B, therefore, levels of B are said to be nested within the levels of factor A. However, the basic principle for estimating variance components has been, and to a large extent still is, that of equating quadratic functions of the observations to their expected values. Obvious candidates for such functions are those of the analysis of variance table. The first papers describing this procedure appear to be those of [7], whose interest was in weights of slubbings from the carding process in the woolen industry, and [28] who analyzed catches of different species in successive hauls of plankton nets. Although [7] and [28] represent both sides of the Atlantic, it appears that major developments in variance components estimation subsequently took place mainly in the U. S. A. An exception to this was [11], dealing with nested classifications, and then came [6] concerned with randomized complete blocks, and [24] dealing with approximate sampling distributions of estimated variance components. This was followed by [9] who put a firm foundation to the distinction between the fixed effects model (Model I) and the random effects model (Model II), a distinction which [29] later took great exception to.

Research paper thumbnail of 4-On-the-Efficiency-of-Some.pdf

This work examines the Relative Efficiency ( ) of some selected Randomized Complete Block Designs... more This work examines the Relative Efficiency ( ) of some selected Randomized Complete Block Designs ( ).The efficiency of the selected designs showed that design was the most preferable having Mean Square Error of 6.14, followed by design with Mean Square Error of 9.11 and design with Mean Square Error of 18.08. The results from the pairwise relative efficiency of the selected designs show that ( , ) = 0.34 with the smallest relative efficiency value and ( , ) = 2.94 with the largest relative efficiency value. We recommended design as the best design for this particular problem since its mean square error remains the smallest.

Research paper thumbnail of Of Students Academic Performance Rates Using Artificial Neural Networks (ANNs

A model based on the multilayer perception algorithm was programmed. The result from the test dat... more A model based on the multilayer perception algorithm was programmed. The result from the test data evaluation showed that the programmed Artificial Neural Network model was able to correctly predict and classify the performance of students with Mean Correct Classification Rate CCR of 97.07%.