[R-sig-Geo] Multiple time-series analysis

[Tom Higginbottom]

Hello,

I am attempting to implement three trend techniques: linear, Mann Kendall
and Theil Senn, on a raster brick consisting of 20 rasters derived from some
statistical analysis.

I have complete the MK and linear trends, although the MK values seem
bizares as there are no negaitive slopes which would be expected. However, I
can not figure out how to implement a Theil Senn trend.

Any help would be much appreciated.

My code so far is below, not i am still new to r and have pulled section of
coding from mailing posts so it may not be the most refined method


#Generate the raster brick
list <- list.files(path ="G:\\gwr_rst\\", pattern = "*.rst$", all.files =
FALSE, full.names = TRUE, recursive = FALSE, ignore.case =
FALSE,include.dirs = FALSE, no.. = FALSE)

stack <- stack(lapply(list, raster))

brick <-brick(stack_80)

#Trend no breaks using the greenbrown package
Linear<- TrendRaster(Brick, start=c(1983, 1),  method="AAT", breaks=0)

#Gridded data Mann Kendall trend 

> myraster.mk <- calc(brick,function(x){MannKendall(rev(x))$sl}) 




--
View this message in context: http://r-sig-geo.2731867.n2.nabble.com/Multiple-time-series-analysis-tp7584485.html
Sent from the R-sig-geo mailing list archive at Nabble.com.