Dear All, I am working on a time series of hourly river flow measurements from 2000 - 2003 and have been trying to compute the extremum index from the original series as well as a new series from sampling with replacement . The extremum Index produced from the original data series looks fine but sadly, a strange value of index is given from the sampling. The flow measurements are in range of 0.1 to 35m3/s and are stored in data frame with the codes and results are as follows: ########################## ### Codes ### attach(data) # Estimates of extremum index library(evd) library(boot) set.seed(101) threshold <- 15 ej <- exi(data$flow, u=threshold) # Extremum Index computed from the original series of river flow ei <- exi(sample(data$flow, replace=T), u=threshold) # Extremum Index computed from sampling of the original flow series with replacement ei; ej mean(ei); mean(ej) ####################### #### Results #### ei = 0.4 ej = 1 mean(ei) = 1.023 mean(ej) = 1.014 ####################### I would expect that the results for both index should be slightly similar as the features of both original & sampling series are more or less the same as highlighted by their sample mean and so, could anybody please advice me on this matter? Thank you, Fir [[alternative HTML version deleted]]
On Jul 8, 2011, at 1:44 PM, FMH wrote:> Dear All, > > I am working on a time series of hourly river flow measurements from > 2000 - 2003 and have been trying to compute the extremum index from > the original series as well as a new series from sampling with > replacement . The extremum Index produced from the original data > series looks fine but sadly, a strange value of index is given from > the sampling. The flow measurements are in range of 0.1 to 35m3/s > and are stored in data frame with the codes and results are as > follows: > > ########################## > > ### Codes ### > attach(data)Nope, no data. Consult the Posting Guide. See link below.> Fir > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.David Winsemius, MD West Hartford, CT