Journal of Statistical Modeling & Analytics (JOSMA)

An ARIMA Modeling of the Exchange Rate of the Nigerian Naira to the US Dollar with Covid-19 Pandemic Event in Focus

EMMANUEL ODUNTAN — 2024-05-30

We modelled the average monthly exchange rate between the Nigerian Naira and the USDollar using univariate time series analysis. With a data set covering January 2002 to June 2022, we divided the data set into three distinct periods, vis-à-vis; the Pre COVID-19 pandemic period, the COVID-19 period, and the Combined period (combination of both pre-COVID-19 and COVID-19 periods). This was done with a view to examining the effect of COVID-19 pandemic on the generation mechanism of the Naira/USDollar exchange rate. The Naira/USDollar exchange rate data was found to be non-stationary and was appropriately differenced to attain stationarity. Subsequently, we fitted ARIMA models for each of the three scenarios using Box Jenkins methodology. Finally, the estimated models were used for forecasting. While the estimated models for both the Pre COVID-19 and the Combined periods yielded forecast values that suggest future depreciation of the Naira against the USDollar, that of the COVID-19 period yielded forecast values that suggest future appreciation of the Naira against the USDollar.

Boosting Cancer Dataset Performance with Mutual Information-Based Feature Prioritization

Fung Yuen Chin — 2024-05-30

In the field of statistical modelling, mutual information is a crucial and common concept, suitable for tasks such as selecting the most important features or classifying data into different categories. Feature selection addresses the challenge of high-dimensional data in building effective predictive models by identifying relevant attributes while mitigating the curse of dimensionality. Previous studies have benchmarked the effectiveness of statistical models against established results. To enhance this, a new benchmark method is proposed, exploiting ranking features via mutual information scores. Mutual information score is used to understand the relationship between underlying data and variables. The performance of the classification depends on its information content, which directly affects the performance of the statistical model. The technique simultaneously determines the optimal feature quantity to guide the feature selection process. The validation of these selected features is conducted through Z-score graphs. Experimental results show that this method can identify feature subsets better than using the full features. This advance promises to improve cancer analysis, enabling more sophisticated diagnostic and prognostic methods.

Multicollinearity in Binomial Regression: A comparison between CERES and PR Plots for detection

Nasir Saleem saleem — 2024-05-30

For identification of multicollinearity, residuals are most common tool in linear regression model, but a limited literature is available which describe this situation in case of GLM. Binomial regression model has extensive applicability in analyzing with heart disease and many other types of data. Here, we have offered a comparison between CERES and PR plots in BRM to detect the multicollinearity problem. At first, we have developed a comparison tool and then apply them to real-world and simulated data. We examine and compare these plots on the detection of a possible multicollinearity separately and observe that the performance of CERES plot is better than compare to the PR plots.

On the Type I Half Logistic Lomax Distribution with applications

Bashiru Omeiza Sule — 2024-05-30

In this study, a new four-parameter lifetime distribution called the Topp Leone exponentiated Lomax distribution was introduced. Expansion of density for the probability distribution function and cumulative density function was done from which some of the properties of the new model were derived. Some mathematical properties of the distribution such as the moments, moment generating function, quantile function, survival function and hazard function were presented. The probability density function of the maximum and minimum order statistics was also derived and studied. Estimation of the parameters by maximum likelihood method was discussed and used to estimate the unknown parameters of the distribution. Four real life data sets were used to show the ﬁt and ﬂexibility of the new distribution over some existing lifetime distributions in literature and the results show that the new distribution ﬁts better in the four data sets considered.