•   •   •   •

Download the Paper (594 K, PDF file) - January 29, 2007 Tell a colleague about it. Printing Tips: Select 'print as image' in the Acrobat print dialog if you have trouble printing.

ABSTRACT:
This paper describes a method of kernel regression that can be used to investigate the relationship between a directional explanatory variable and a real-valued response variable. Cross-validation and bootstrap methods for obtaining sensible bandwidths and standard error estimates are also described. The proposed method is applied to wildfire and meteorological data from Los Angeles County, California, with the goal of summarizing and quantifying the impact of wind direction on the total area burned per day in wildfires. The results confirm that winds blowing from the NorthEast and East are associated with significantly higher burn areas than winds from other directions and that the daily burn area on days with winds from the NorthEast is about 4.7 times that associated with the winds from the SouthWest.

SUGGESTED CITATION:
Haiyong Xu and Frederic P. Schoenberg, "Kernel regression of directional data with application to wind and wildfire data in Los Angeles County, California" (January 29, 2007). Department of Statistics, UCLA. Department of Statistics Papers. Paper 2007010110.
http://repositories.cdlib.org/uclastat/papers/2007010110