Analysis of different Mathematical Models for Different Case Studies Using Statistical Fitting

TL;DR

This paper evaluates various statistical curve fitting models across three case studies—population dynamics, building temperature, and market prices—highlighting the importance of selecting appropriate functions for accurate predictions.

Contribution

It systematically compares different mathematical models for diverse case studies, identifying the most effective functions based on error criteria.

Findings

Fractional exponential functions best fit population data.

Sinusoidal functions with three terms suit temperature and market data.

Model selection depends on the specific nature of the case study.

Abstract

Curve fitting is a fundamental technique in engineering and scientific research, serving as a critical tool for extracting insights from data. This study explores the application of various statistical equations to estimate outcomes in three distinct case studies: population dynamics, temperature variations within buildings, and market equilibrium prices. The efficacy of each fitting is evaluated through rigorous error criteria, including Sum of Squares Error (SSE), R-squared (R), Degrees of Freedom Error (DFE), Adjusted R-squared (Adj. R), and Root Mean Square Error (RMSE). Our findings reveal that the selection of mathematical functions and the order of equations are contingent upon the specific nature of the model being analyzed. In the first case study concerning population dynamics, a fractional exponential function emerges as the optimal equation. Conversely, the second and third case studies, which focus on temperature fluctuations in buildings and market price equilibrium, respectively, find the best fit with a sinusoidal function employing three terms. Additionally, we compare the fitted curves and empirical equations with experimental data from the literature to comprehensively assess their predictive accuracy. By establishing the most suitable mathematical models for diverse case studies, researchers and practitioners can enhance their ability to make accurate predictions and informed decisions.