Open Access Original Research Article

Research on Operational Risk Management in First Bank of Nigeria (FBN Bank)

Babatunde Ayomikun Seun, Zhang Diping, Babalola Emmanuel Olusola

Asian Journal of Probability and Statistics, Page 1-27
DOI: 10.9734/ajpas/2022/v17i230416

The FBN bank is the oldest and largest of the twenty-four banks operating in the Nigerian economy. The objective of this study is to assess the impact of Operational Risk Management on FBN Banks. Past literatures were reviewed and theoretical frameworks such as the extreme-value theory and risk theory of profit were adopted to support the study. The research adopted both qualitative and quantitative techniques, and the data for the study was employed from primary and secondary sources. Primary data questionnaires were distributed to 60 bank workers, but Only 50 surveys were returned from the served respondents, and the analysis was focused on those 50.  Eventually, the respondents' responses were analyzed using simple percentages. Moreover, the secondary data were derived from the sampling deposit money institutions audited and publicly available financial statements of first bank of Nigeria. Following, the results were analyzed based on time series basis from 1999 to 2020 using regression estimates. The investigation indicated that operational risk and credit risk have a greater impact on FBN banking operations than market risk. Fraud and forgeries also have a negative impact on banking operations. However, fraud and forgery risk, operational risk, credit risk, and system risk abound in FBN banking operations, all of which must be managed effectively to improve bank performance and stability. Deductively from the survey analyses, the FBN banks' risk management procedures have effectively minimized the different risks that FBN banks face. Nevertheless, a case study was carried out and these necessitates the design of a risk indicator system that will further help first bank of Nigeria to map and curb their operational risks.

The regression result revealed that operational risk management had a considerable impact on FBN banks' performance measures in Nigeria. Especially the Ratio of Non-Performing Loans to Total Loans (BRNPL), and Ratio of Cost to Income (BROCI) has a negative significant impact on the financial stability of FBN banks in Nigeria as measured by Return on Equity (ROE).

Open Access Original Research Article

Power Series Generalised Power Weibull Class of Distributions

Anuwoje Ida L. Abonongo, Albert Luguterah, Suleman Nasiru

Asian Journal of Probability and Statistics, Page 28-51
DOI: 10.9734/ajpas/2022/v17i230417

The power series generalised power Weibull class of distributions were developed in this study by compounding the power series family of distributions and the generalised power Weibull distribution. The statistical properties of this new class were derived. Maximum likelihood parameter estimators were derived for the parameters of the power series generalised power Weibull class of distributions. Four sub-families of distributions were developed from the power series generalised power Weibull class of distribution; the generalised power Weibull geometric distribution, generalised power Weibull Poisson distribution, generalised power Weibull binomial distribution and the generalised power Weibull logarithmic distribution. The hazard rate and probability density function plots of the four sub families of distributions showed that, they can model both monotonic and non-monotonic lifetime data. Monte Carlo simulations performed on these sub-distributions showed that, their estimators were consistent estimators. Application of these sub-distributions to failure data from air conditioning system of an aircraft showed that, the generalised power Weibull geometric distribution provides a better fit to the data. Also, the generalised power Weibull Poisson distribution provides a better fit to the data on service times of 63 aircraft.

Open Access Original Research Article

Impact Assessment of Gap on Nigerian Crude Oil Production: A Box-Tiao Intervention Approach

Imo U. Moffat, Elisha J. Inyang

Asian Journal of Probability and Statistics, Page 52-60
DOI: 10.9734/ajpas/2022/v17i230419

The Nigerian disarmament, demobilization and reintegration programme known as the Amnesty Programme have become common practice in countries facing violent conflict. A time-series model for monthly crude oil production was developed to examine the effect of Government’s Amnesty Programme (GAP) introduced in August 2009. The data used in this study are the monthly crude oil production spanning from January 1999 to December 2020. From results of model estimation, we found that the intervention due to the Amnesty programme had no impact on crude oil production since the null hypothesis that is 0 was retained. Factors suggested to account for this development includes: Crack in deal, as Nigerian Government found it difficult to fund the programme due to incessant fall in oil price and emergence of new groups of militant. The renewed instability and violent attacks on oil facilities have resulted in a serious reduction of crude oil production. Thus, government should consider tackling wider socio-economic grievances in the country’s oil production region in other to bring stability, which will lead to increase in oil production.

Open Access Original Research Article

Initialization and Estimation of Weights and Bias using Bayesian Technique

Oryiema Robert, David Angwenyi, Nyogesa Achiles, Jacob On'gala

Asian Journal of Probability and Statistics, Page 61-85
DOI: 10.9734/ajpas/2022/v17i230420

Researches on artificial neural network models have shown that the method used to initialize and estimate weights and bias always determines the rate at which the network will converge and how efficient the model will perform. Although there are several methods that can be used to initialize and estimate network weights and bias, Bayesian approach is widely considered as a more efficient method for modelling artificial neural networks because it can easily compute the inverses of covariance matrices with high dimension that otherwise are computationally expensive. This study has developed a new filter, the First order Extended Ensemble Filter(FoEEF),that applies numerical solution in solving the inverses of high dimensional covariance matrices from I ^ to ′ s stochastic state-space dynamical models. The research applies the new FoEEF Filter to initializes and estimates the weights and bias of artificial neural network. Comparison on the performance of FoEEF filter in estimating weights and bias are done against the performance of EKF on function estimations by using two different functions, sin(x)+Q function and 3sin(x)3− 3sin(x)2 − sin(x) + 1 function, within 18 and 22 epochs. Emphasis of the performance is placed on the rate at which the models converge. This study gauges the performance by determining the minimum value of the mean square error produced from each epoch and average mean square error minimum value.The outcome from the study showed that FoEEF filter had the lowest value of mean square error and average mean square error which was achieved with the least number of epochs compared to EKF filter. This study concluded that the new FoEEF performed better than EKF and is a more suitable filter for that can be used for initialization and estimation of weights and bias in artificial neural network.

Open Access Original Research Article

Extreme Value Theory Modeling of Geochemical Anomalies: Block Maxima Approach

Emmanuel Ayitey, Christiana C. Nyarko, Henry Otoo, Micheal Affam

Asian Journal of Probability and Statistics, Page 86-95
DOI: 10.9734/ajpas/2022/v17i230421

Mineral shortages can be avoided if the mineral industry accurately predicts mineral deposits, which is critical given the importance of minerals in Ghana's economy. The goal of this dissertation was to use the block maxima (BM) approach of Extreme Value Theory (EVT) to accurately predict gold (Au) concentration and the time period of occurrence of these geochemical anomalies in Ghana's Wassa-Amenfi region. The information was based on a time series of daily gold concentrations collected between 2010 and 2018 by Ghana's geology and survey department. The shape parameter estimates from the analysis indicated that the Fréchet family of GEV distributions was a good fit for the dataset. The GEV model was used to forecast the occurrence of anomalies every 2, 5, 10, 20, 50, and 100 years. According to the findings, an extreme Au of 31.06 was expected to occur once every 5 years in Wassa-Amenfi.