Comparison of six generalized linear models for occurrence of lightning-induced fires in northern Daxing’an Mountains, China

• Published in: Journal of forestry research Vol. 27; no. 2; pp. 379 - 388
• Main Authors: Guo, Futao, Wang, Guangyu, Innes, John L, Ma, Zhihai, Liu, Aiqin, Lin, Yurui
• Format: Journal Article
• Language: English
• Published: Harbin Northeast Forestry University 01.04.2016; Springer; College of Forestry, Fujian Agriculture and Forestry University, Fuzhou 350002, China; Sustainable Forest Management Laboratory, Faculty of Forestry, University of British Columbia, 2424 Main Mall,Vancouver, BC V6T 1Z4, Canada%Sustainable Forest Management Laboratory, Faculty of Forestry, University of British Columbia, 2424 Main Mall,Vancouver, BC V6T 1Z4, Canada%Department of Medicine, University of Calgary, 3280Hospital Drive NW, Calgary, AB T2N 4Z6, Canada%College of Forestry, Fujian Agriculture and Forestry University, Fuzhou 350002, China%College of Computer and Information Science, Fujian Agriculture and Forestry University, Fuzhou 350002, China
• Subjects: Analysis; Biomedical and Life Sciences; Forest fires; Forestry; Life Sciences; linear models; meteorological parameters; mountains; Original Paper; prediction; variance; Weather; China; Zero-inflated Poisson (ZIP); Poisson hurdle (PH); Poisson; Vuong test; Zero-inflated negative binomial (ZINB); Negative binomial (NB); Negative binomial hurdle (NBH); Likelihood ratio test (LRT)
• ISSN: 1007-662X; 1993-0607
• DOI: 10.1007/s11676-015-0176-z

The occurrence of lightning-induced forest fires during a time period is count data featuring over-dispersion (i.e., variance is larger than mean) and a high frequency of zero counts. In this study, we used six generalized linear models to examine the relationship between the occurrence of lightning-induced forest fires and meteorological factors in the Northern Daxing’an Mountains of China. The six models included Poisson, negative binomial (NB), zero-inflated Poisson (ZIP), zero-inflated negative binomial (ZINB), Poisson hurdle (PH), and negative binomial hurdle (NBH) models. Goodness-of-fit was compared and tested among the six models using Akaike information criterion (AIC), sum of squared errors, likelihood ratio test, and Vuong test. The predictive performance of the models was assessed and compared using independent validation data by the data-splitting method. Based on the model AIC, the ZINB model best fitted the fire occurrence data, followed by (in order of smaller AIC) NBH, ZIP, NB, PH, and Poisson models. The ZINB model was also best for predicting either zero counts or positive counts (≥1). The two Hurdle models (PH and NBH) were better than ZIP, Poisson, and NB models for predicting positive counts, but worse than these three models for predicting zero counts. Thus, the ZINB model was the first choice for modeling the occurrence of lightning-induced forest fires in this study, which implied that the excessive zero counts of lightning-induced fires came from both structure and sampling zeros.