The availability of player performance data has significantly enhanced the scope and depth of quantitative analysis in recent professional football; however, position-specific statistical investigations, especially for specialized roles such as wide midfielders, remain relatively unavailable. This study addresses this gap by carefully studying these correlations and picking out relevant determinants of players' performance for left and right midfield positions using adequate multivariate statistical techniques. A comprehensive dataset comprising physical, technical, and mental attributes of modern-day professional players occupying these wide midfield positions was collected and subjected to rigorous statistical procedures. The study also deployed Pearson correlation analysis to explore linear relationships between individual player attributes and overall performance ratings. Canonical correlation analysis (CCA) was used to determine the relative contribution of each attribute, as well as principal component analysis (PCA), to uncover latent performance dimensions and reduce redundancy among some of the highly correlated variables. The empirical results show that physical and mental attributes, especially ball control, dribbling, vision, short passing, and composure, show strong positive correlations with the general performance ratings of the players. In contrast, physical attributes such as sprint speed and acceleration show comparatively weaker and less consistent relationships. Furthermore, the presence of strong intercorrelations among technical variables suggests substantial overlap among performance indicators, therefore, justifying the use of dimensionality-reduction techniques in the study. Finally, the study highlights the importance of technical proficiency and decision-making ability in determining the effectiveness of wide midfielders. These revelations provide valuable empirical support for data-driven approaches to identification of talent, player development, tactical improvements and maximization, and scouting strategies in modern football analytics.
This study examines the distributional behavior of Nigeria’s Gross Domestic Product (GDP) using the Generalized Inverted Exponential Distribution (GIED) as a flexible parametric framework for modeling macroeconomic data. GDP series in developing economies are often characterized by skewness and departures from normality, which limits the adequacy of conventional distributions such as the Log-Normal and Pareto models in capturing their empirical properties. This motivates the use of more flexible alternatives for improved distributional modeling. The analysis is based on Nigeria’s GDP data spanning 1981–2022. Model parameters are estimated using Maximum Likelihood Estimation (MLE), while model adequacy is evaluated through log-likelihood values, Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), Kolmogorov–Smirnov (K–S) statistics, and graphical diagnostic tools. The results indicate that the GIED provides a superior overall fit compared to the competing distributions, as reflected in lower information criteria values and improved goodness-of-fit statistics, highlighting its flexibility in capturing skewed and asymmetric features of GDP data. In addition, a simple linear time-trend regression model is employed to assess the temporal evolution of GDP. The estimated time coefficient is positive and statistically significant, and its stability is confirmed through bootstrap resampling and Bayesian robustness checks, indicating consistency across estimation approaches. Overall, the findings suggest that the GIED is a suitable and flexible distribution for modeling GDP in contexts characterized by non-normality, while the regression results provide complementary evidence of a sustained upward trend in Nigeria’s GDP over the study period. This supports the usefulness of combining distributional and regression approaches in macroeconomic data analysis.
In general, thermal analysis is a very convenient way to study the kinetics of thermally activated solid reactions, and experiments involving thermally activated solid reactions typically occur under nonisothermal conditions. The experimental data analysis obtained under non-isothermal conditions includes a temperature integral, also called the Arrhenius integral, which is one of the integrals that belong to many interesting integrals that are important in engineering and have no analytical solution. There are many different approximations that can be applied to the processing of thermogravimetric analysis data, since there is no standard way to calculate the temperature integral. Many approximations of the temperature integral that are important to use determine the kinetic parameters, especially the activation energy, which are typically divided into two categories: exponential and rational approximations. In order to evaluate the accuracy of various approximations of the temperature integral, we consider several certain continuous intervals. When choosing an approximation of the temperature integral needed to analyze the experimental data, it is necessary to analyze the accuracy at different temperature intervals of the approximation and use the appropriate one. We present new rational, irrational and continued fractional approximations together with approximations of the temperature integral presented in several literatures and calculate the relative errors of their activation energies.
Regression analysis is a core analytical tool widely employed across diverse domains for predicting continuous outcomes, serving as a cornerstone of statistical inference and machine learning applications ranging from economic trend forecasting to healthcare risk assessment and real estate valuation. Choosing an effective regression technique is critical for accurate predictions, yet a daunting challenge for non-experts due to the wide variety of methods, each with distinct assumptions, tuning requirements and applicability boundaries. To address this dilemma, this study conducts a rigorous empirical comparison of five popular regression techniques—Ordinary Least Squares (OLS), Ridge regression, Lasso regression, Elastic Net, and Polynomial regression—applied to house price prediction using two benchmark datasets: the classic Boston Housing dataset and the comprehensive California Housing dataset. A multi-dimensional evaluation framework was adopted, including quantitative metrics (Mean Squared Error (MSE) and coefficient of determination () and qualitative diagnostics (residual analysis and Quantile-Quantile (QQ) plots) to assess prediction accuracy and error distribution. Results indicate that Polynomial regression consistently achieves superior performance across both datasets, highlighting its effectiveness in capturing the complex nonlinear relationships inherent in housing data. Ridge, Lasso, and Elastic Net provide comparable but lower performance, with strengths in mitigating multicollinearity rather than enhancing nonlinear fitting. OLS yields acceptable baseline results but less robust performance when confronted with real-world nonlinearities. These findings offer clear practical guidance for non-experts seeking reliable “out-of-the-box” regression techniques, and contribute valuable insights to assist practitioners in model selection for real-world predictive tasks without extensive tuning.
Grid-connected solar photovoltaic (PV) systems play a critical role in addressing Nigeria’s persistent electricity shortages. However, investment decisions in these systems are often constrained by high tariffs, substantial upfront costs, and significant policy and market uncertainties. Such uncertainties are not well captured by conventional valuation techniques, particularly the Net Present Value (NPV) method, which assumes deterministic cash flows and overlooks managerial flexibility. To address these limitations, this study applies Real Options Analysis (ROA) to evaluate a 16.8 kWp solar PV system installed in a Nigerian government organisation, explicitly quantifying the value of deferring the investment under uncertainty using a binomial lattice framework. Project volatility (σ = 0.3098) was estimated through Monte Carlo simulation incorporating stochastic variations in electricity tariffs, energy yield, and operating costs. The results indicate that while the traditional NPV is positive at N=17.24 million, the option to defer adds an additional N=25.78 million, yielding an Expanded NPV (ENPV) of N=43.01 million-representing a 149% improvement over the static valuation. This substantial flexibility premium demonstrates that exclusive reliance on deterministic NPV significantly undervalues renewable energy investments in volatile environments. The findings highlight the importance of integrating uncertainty and strategic timing into project appraisal, especially in developing countries like Nigeria where regulatory, financial, and market risks are pronounced. The study offers a replicable decision-support framework for investors and underscores the need for stable and supportive renewable energy policies to enhance the adoption of solar PV technologies.
