Portal de Periódicos da CAPES

Comparison of Collinearity Indices for Linear Models in Agricultural Trials

2024; Volume: 24; Issue: 2 Linguagem: Inglês

10.3844/ojbsci.2024.195.207

2410-8561

Danny Villegas Rivas, José Manuel Palacios Sánchez, Cristina A. Alzamora Rivero, Carlos M. Franco Del Carpio, César Osorio Carrera, Martín Grados Vásquez, Luis Ramírez Calderón, Luís Edgardo Cruz Salinas, Karin Ponce Rojas, Liliana Correa Rojas, José Jorge Rodríguez Figueroa, Felicia Lelia Cáceres Narrea, Delia A. Saravia Pachas, Arrieta Benoutt Felipe, Arturo N. Neyra Flores, Pedro E. Zata Pupuche, Carlos Fabián Falcón, Yolanda Maribel Mercedes Chipana Fernández, Marilú Trinidad Flores Lezama, Asunción Ramón Lezcano Tello, Pablo Valentino Aguilar Chávez, Víctor Hugo Fernández Rosas, Francisco Alejandro Espinoza Polo, Gaby Esther Chunga Pingo, Mercy Carolina Merejildo Vera, Carlos Alfredo Cerna Muñóz, Luis Orlando Miranda Díaz, Miguel Ángel Hernández López, Martín Desiderio Vejarano Campos, Erick Delgado Bazán, Zadith Garrido Campaña, José Paredes Carranza, Leyli J. Aguilar Ventura, Graciela M. Monroy Correa, Ruth A. Chicana Becerra, Jhonny Richard Rodriguez-Barboza, Rafael Damián Villón Prieto, Claudia Rosalía Villón Prieto, Mariella M. Quipas Bellizza, Fernando Emilio Escudero Vílchez, Silvia Liliana Salazar Llerena,

Advanced Statistical Methods and Models

The deleterious consequences of collinearity in linear regression on the precision of estimators of regression coefficients and the interpretability of the fitted model are widely recognized. In this study, we compare several methodologies for assessing collinearity in linear models and explore the effect of outliers on collinearity. The robustness of collinearity measures (individual and overall) is validated through two detailed Monte Carlo simulation study which also considers the effect of outliers on collinearity indices. The methods are illustrated with two real-world agricultural and fish morphology l data sets to show potential applications. The results do not provide any evidence for an effect from outliers on collinearity identification using the collinearity indices (individual and overall). The FG and Fj collinearity indices more robust as both sample size and collinearity degree increase. The VIF (individual measure) had a better performance on the fitted model with a greater number of parameters.