bestglm (version 0.13)

Fires: Forest fires in Montesinho natural park. Standardized inputs.

Description

The forest fire data were collected during January 2000 to December 2003 for fires in the Montesinho natural park located in the northeast region of Portugal. The response variable of interest was area burned in ha. When the area burned as less than one-tenth of a hectare, the response variable as set to zero. In all there were 517 fires and 247 of them recorded as zero. The region was divided into a 10-by-10 grid with coordinates X and Y running from 1 to 9. The categorical variable xyarea indicates the region in this grid for the fire.

Usage

data(Fires)

source

http://archive.ics.uci.edu/ml/datasets/Forest+Fires

Details

The original data may be found at the website below as well as an analysis. The quantitative variables in this dataset have been standardized. For convenience, the original data is provided in MontesinhoFires.

References

P. Cortez and A. Morais, 2007. A Data Mining Approach to Predict Forest Fires using Meteorological Data. In J. Neves, M. F. Santos and J. Machado Eds., New Trends in Artificial Intelligence, Proceedings of the 13th EPIA 2007 - Portuguese Conference on Artificial Intelligence, December, Guimaraes, Portugal, pp. 512-523, 2007.

See Also

MontesinhoFires

Examples

data(Fires)
names(Fires)
#Fit full model and compare with anova and best AIC subset
#Fit full model
bestglm(Fires, IC="BICq", t=1)
#ANOVA for xyarea is significant at 1.1%.
summary(aov(lburned~xyarea, data=Fires))
bestglm(Fires, IC="AIC")