Analyzing the mahakam river water quality using the geographically weighted panel regression model

This study discusses the geographically weighted panel regression (GWPR) model. GWPR is an extension of geographically weighted regression model, designed for spatially heterogeneous panel data. In this study, GWPR model is applied to panel data on biochemical oxygen demand (BOD) in Mahakam River water 2022-2024. The model is estimated at each spatial location using a fixed effects model (FEM) as the global model, with temporal effects addressed through a demeaning transformation. All statistical analyses and spatial processing are conducted using R software, GNU Octave, QGIS, and Google Earth. This study aims to map factors influencing Mahakam River water BOD using GWPR model. The results indicate that GWPR outperforms FEM, with AIC = -60.6419, <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:msup><mml:mrow><mml:mi>R</mml:mi></mml:mrow> <mml:mn>2</mml:mn></mml:msup> <mml:mo>=</mml:mo> <mml:mn>80.321</mml:mn> <mml:mo>%</mml:mo></mml:mrow> </mml:math> , and root mean square error of 0.7122. The factors influencing BOD include temperature, water pH, color degree, nitrate, ammonia, total suspended solids, and sulfate.•<i>We present a GWPR model using FEM as global model, applied to the spatially heterogeneous panel data, namely demeaned Mahakam River water BOD data 2022-2024.</i>•<i>The mapping of factors influencing BOD is analyzed locally using GWPR model.</i>•<i>The optimal adaptive bandwidth is determined using Akaike Information Criterion, and model goodness-of-fit is evaluated using the coefficient of determination and root mean square error.</i>