| 1 |
Author(s):
Nwakuya Maureen T., Nkwocha Chibueze C. .
Page No : 1-11
|
Quantile Regression for Count Data as a Robust Alternative to Negative Binomial Regression
Abstract
The study investigated the robustness of Quantile regression of count data over negative binomial regression, when there is overdispersion and presence of outlier. The study made use of a complete data and the data with 30% missing data which was imputed using Multiple Imputation by Chain Equation (MICE) in R and also an outlier was injected into the data during imputation of missing values. The Quantile Regression and Negative Binomial Regression estimates were compared and their model fits were also compared. Results showed that the quantile regression for count data provided a better model estimate with both complete data and data with multiple imputed value with comparison to the negative binomial regression in terms of AIC, BIC RMSE and MSE. Hence, Quantile Regression is better than the negative binomial regression when the researcher is interested in the effect of the independent variable on different points of the distribution of the response variable and when there is overdispersion and presence of an outlier.
| 2 |
Author(s):
Christogonus Ifeanyichukwu Ugoh, Lamin B. Jammeh, Mark Nwachukwu Ugo, Emwinloghosa Kenneth Guobadia, Chimezie Stanley Ngene.
Page No : 12-26
|
Modeling and Forecasting Nigerian Naira/US Dollar and The Gambian Dalasi/US Dollar Exchange Rates: A Comparative Study
Abstract
This paper compares the predictive performance of time series forecast methods on the Nigerian Naira/US Dollar (NGN/USD) and The Gambian Dalasi/US Dollar (GMD/USD) exchange rates. The forecast methods—Autoregressive Integrated Moving Average (ARIMA), Simple Exponential Smoothing (SES), Holt’s Linear Trend, and Damped Holt—were applied to the annual Nigerian Naira and Gambian Dalasi against the US Dollar for the period 1960–2020. The best model for forecasting exchange rates in both countries was selected based on Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Mean Absolute Scaled Error (MASE). The findings in this study revealed that both Naira/US Dollar and Gambian Dalasi/US Dollar exchange rate distributions are positively skewed and ARIMA (0,2,2) model was selected as the most appropriate model for forecasting both exchange rates. The results also showed that by 2030, the Nigerian Naira/US Dollar exchange rate will rise by 37.06 percent while the Gambian Dalasi/US Dollar will rise by 23.18 percent. This study suggests that both countries should adopt tighter fiscal, monetary, and supply-side policies.
| 3 |
Author(s):
Jophet Ewere Okoh, Udochukwu Victor Echebiri, Emwinloghosa Kenneth Guobadia, Akpome Jennifer Nomuoja, Raphael Michael Ugochukwu.
Page No : 27-45
|
On Intractability in G-Class of Lifetime Probability Distributions: Properties and Applications
Abstract
This research seeks to project the relevance of incorporating intractability in G-Class probability model development. Juchez distribution represents intractability among the conventional tractable distributions, as comparatively studied in their different G-Class dimensions. Some properties were derived; although nonintegrable properties really were a confirmation note to the intractable nature of the distribution. The L-shape hazard function as a special feature across the G-Class derived distributions suggests that the distribution is fit to model outcomes that do not wear out as time or cycle elapses. Finally, the performance comparison reveals that intractability could be embraced in the development of probability models, as the derived distributions showed to be a better fit than the existing G-class baseline-tractable distributions.
| 4 |
Author(s):
Christogonus Ifeanyichukwu Ugoh, Chukwuemeka Thomas Onyia, Jophet Ewere Okoh, Pamela Owamagbe Omoruyi.
Page No : 46-55
|
Is Single Forecast Method Better than Combined Forecast Method?
Abstract
Many studies have been done to prove that combining forecast methods gives a better predictive performance relative to individual forecasts. This paper compared the single forecast method and the combined methods in predicting time series data. The study used annual oil revenue for the period 1981–2019 from the Central Bank of Nigeria (CBN), which were divided into two sets: the Training Set (TS) which covered the period 1981–2010 and the Test Set (VS) which covered 2011–2019. The study adopted autoregressive integrated moving average (ARIMA), simple exponential smoothing (SES), and Holt’s linear trend (Holt) as the individual forecast methods; it also adopted outperformance of forecasts (OPF) and weighted mean (WM) as weight selection methods. The forecast methods were applied to the Training Set after which they were combined. Two combined methods CM1 (ARIMA + SES) and CM2 (ARIMA + SES + Holt) were obtained. The result of this study showed that simple exponential smoothing (SES) as an individual forecast method is better and less risky than the combined methods for forecasting time series.
| 5 |
Author(s):
Seun Adebanjo, Pius Sibeate.
Page No : 56-69
|
Probability Modeling of Exchange Rate Fluctuation in Nigeria
Abstract
The Naira's value has continued fluctuating in comparison with other currencies due to its depreciation. Because of this fluctuation, returns are difficult to forecast. The core objective of this study is to find a unified probability distribution for modelling the exchange rate in Nigeria and this will contribute immensely to the existing body of knowledge. The continuous nonnegative exchange rate data from 1970 to 2021 was used for this research paper. Previous studies have demonstrated different probability distributions from others in the real sense. Therefore, the selection of appropriate probability distributions is of great importance. This study adopted ten continuous probability distributions. The graph of the probability density function and Chi-square goodness of fit statistics show that the probability distributions fit the exchange rate data. Meanwhile, the log-likelihood value and the AIC show that Fatigue Life (3P) distribution is best done compared with fitted probability models. Therefore, the Fatigue Life (3P) is Nigeria's unified probability distribution for the modelling exchange rate.
| 6 |
Author(s):
Adebanjo Seun Adebowale, Sibeate Pius, Oladapo Ifeoluwa David, Olugbode Morufu Adeoye, Ehinmilorin Elisa.
Page No : 70-87
|
Determinants of Natural Gas Consumption in Europe: An Empirical Analysis
Abstract
Natural gas consumption is a significant issue in European countries due to the Russia-Ukraine war crisis. Natural gas is very important both for household and commercial purposes. The primary objective of this study is to investigate the relationship between natural gas consumption and its determinants. Panel data collected from Eurostat and World Bank publications, consisting of five European countries spanning from 2009 to 2022, were used for this study. Panel data analysis, like the panel unit root test, shows that the panel data variables are integrated into order 1. This indicates that the estimators are sufficient since the variables in the panel are stationary. Johansen Fisher's panel cointegration test shows that there is a long-run association between natural gas consumption and its determinants. The Hausman test specified a panel random effect regression model to run the analysis of this paper, and the model indicates that there is a significant relationship between natural gas consumption and its determinants. The Panel regression model further reveals natural gas prices have a negative significant impact on natural gas consumption, which suggests that the consumption of natural gas reduces with an increase in its price. This is the current situation in European countries now following the effect of the Russian-Ukraine war. Besides, correlation analysis was applied and shows a negative and significant relationship between natural gas consumption and natural gas demand. Following the outcome of this research paper, it will be very important for the United Nations and the European Union to swiftly apply a drastic and lasting solution approach to the current Russian-Ukraine war in order to prevent further untold damages that the war could cause to the economy of European Countries.
| 7 |
Author(s):
Ojonubah James Omaiye.
Page No : 88-102
|
Numerical Analysis of Ordinary Differential Equations of Ecological Competing Species Across Diverse Environments
Abstract
In a geographical region, species have their range margins (i.e., the geographic boundaries where species can be found). Several species distribution models have shown that environmental factors (i.e., abiotic factors) and species interactions (i.e., biotic interactions) are responsible for shaping the distributions of species. Yet, most of the models often focus on one of these factors and ignore their joint effects. Consequently, predicting which species will exist and at what range margins is a challenge in ecology. Thus, in this paper, the combined influences of these ecological factors on multi-species community structures are studied. An ordinary differential equations (ODE) model is employed to study multi-species competition interactions across diverse environments. The model is numerically analysed for the range margins of the species and threshold values of competition strength which leads to the presence-absence of species. It is observed that the range margins are influenced by competition between species combined with environmental factors and the threshold values of competition strength correspond to transcritical bifurcation. Depending on the species’ competition strengths, the model exhibits coexistence and exclusion of species, mediated by weak and aggressive biotic interactions, respectively. It is observed that ecologically similar species competitively affect each other more than dissimilar species.
| 8 |
Author(s):
Kelachi P. Enwere, Uchenna P. Ogoke.
Page No : 103-115
|
A Comparative Approach on Bridge and Elastic Net Regressions
Abstract
Machine Learning techniques such as Regression have been developed to investigate associations between risk factor and disease in multivariable analysis. However, multicollinearity amongst explanatory variables becomes a problem which makes interpretation more difficult and degrade the predictability of the model. This study compared Bridge and Elastic Net regressions in handling multicollinearity in multivariable analysis. Wisconsin Diagnostic Breast Cancer data was used for comparison for model fit and in handling multicollinearity between the regression techniques. Comparison were made using MSE, RMSE, R^2, VIF, AIC and BIC for efficiency. Scatter plots was employed to show fitted regression models. The results from the study show that, the Bridge regression performed better in solving the problem of multicollinearity with VIF value of 1.182296 when ???? = 2 compared to Elastic Net regression with a VIF value of 1.204298 respectively. In comparison for best model fit, Bridge regression with ???? = 0.5 performed better with MSE of 11.58667, AIC value of 258.9855 and BIC of 277.2217 respectively. Consequently, we can conclude that both the Bridge and Elastic Net Regressions can be used in handling multicollinearity problems that exist in multivariable regression analysis. Information on machine learning such as this, can help those in the medical fields to improve diagnosis, narrow clinical trials and biopsy to proffer effective treatment.
