Data of forest fire occurrence from Altamirano et al. (2013) as our population, containing 7210 total observations (N), with only 890 cases of fire occurrence (N 1 ) and 6320 cases of non occurrence (N0). The binary variable (Y) is the occurrence of forest fire, where Y equal to 1 denotes occurrence and Y equal to 0 otherwise.
data(forestfire)The data frame contains four variables as follows:
Presence of forest fire (1 yes, 0 no)
Geographic coordinate x.utm
Geographic coordinate y.utm
Exposure (degrees from north)
Elevation (m)
Slope (degrees)
Distance to dirt roads
Distance to cities
Distance to paved roads
Land use classifications according to a polygon
Land use classifications according to a polygon
Minimum temperature of the coldest month
Annual precipitation
Normalized difference infrared index
Normalized difference vegetation index
Minimum temperature of the warmest month
Precipitation of the driest month
Frequency of fires
Percentage of fire frequency
Class for frecuency fire
Class of variable exposure
Class of numerical variable elevation
Class of numerical variable slope
Normalized difference infrared index class
Normalized difference vegetation index class
- Altamirano A, Salas C, Yaitul V, Smith-Ramirez C, Avila A. 2013. Infuencia de la heterogeneidad del paisaje en la ocurrencia de incendios forestales en Chile Central. Revista de Geografia del Norte Grande, 55:157-170, 2013. -Salas-Eljatib C, Fuentes-Ramírez A, Gregoire TG, Altamirano A, Yaitul V. 2018. A study on the effects of unbalanced data when fitting logistic regression models in ecology. Ecological Indicators 85:502-508. tools:::Rd_expr_doi("10.1016/j.ecolind.2017.10.030")
data(forestfire)
head(forestfire)
Run the code above in your browser using DataLab