[R] Formatting time stamps for weibull distribution

kmmoon100 k.moon at student.unimelb.edu.au
Mon Nov 4 08:50:09 CET 2013


Hello everyone,

I have been using bReeze package for weibull distribution analysis.
In this package, there is a function called 'formatTS' for converting time
stamps from string to POSIXlt.
(http://rgm3.lab.nig.ac.jp/RGM/R_rdfile?f=bReeze/man/formatTS.Rd&d=R_CC)
This is a part of process to plot and calculate shape and scale of the
distribution. 
However, this function keeps showing an error message.

ts1 <- formatTS(time.stamp = winddata_ballarat[,1])
1  1383549939 
2  1383549939 
3  1383549939 
4  1383549953 
Error in formatTS(time.stamp = winddata_ballarat[, 1]) : No pattern found

The format of the time series (200000 rows) I have is dd.mm.yyyy hh:mm for
12 years.. such as..

Time	                   Wind speed    Wind direction
25.07.2000 0:00:00	15	20
25.07.2000 0:30:00	11	20
25.07.2000 1:00:00	18	10
25.07.2000 1:30:00	17	10
25.07.2000 2:00:00	22	360
25.07.2000 2:30:00	22	360
25.07.2000 3:00:00	26	360
25.07.2000 3:30:00	28	360
.                                         .             .
.                                         .             .
.                                         .             .
.                                         .             .
.                                         .             .


I have no idea why the function cannot find a pattern..

Surprisingly, when the same data and the same function was used in my
colleague's computer, it worked. So I tried processing in other computers
but failed. 

I tried using strftime function too but result is not reliable, which
indicates there is a mismatch between the time stamp I created and wind
speed.

Could anyone know why it keeps happening and how to solve this problem?
All R codes are described below:

winddata_ballarat <- read.csv("ballarat1.csv",header=TRUE,sep=",")
set2 <- createSet(height=10, v.avg=winddata_ballarat$Wind.speed,
dir.avg=winddata_ballarat$Wind.direction)
ts1 <- formatTS(time.stamp = winddata_ballarat[,1])
kangmin.ballarat <- createMast(time.stamp=ts, set2)
kangmin.ballarat.wb <- weibull(mast=kangmin.ballarat, v.set=1, print=FALSE)
plotWeibull(wb=kangmin.ballarat.wb, show.ak =TRUE)
weibull(mast=kangmin.ballarat, v.set=1)

Thank you.





--
View this message in context: http://r.789695.n4.nabble.com/Formatting-time-stamps-for-weibull-distribution-tp4679666.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list