Hi Eliza To me it seems like that you're not thinking before you messing about with the data before an analysis.
The years with data for 366 days is leap years. It happens every fourth year and the extra day falls on the 29th of februar. I guess it is the results from the dcast function that screws things up to make you believe that it's day number 366. The best thing to do is to do your analysis on the complete data with some missing values for February 29th between leap years. Or you can discard the leap year day for leap years and do the analysis for all years of 365 days. What is the rationale by imputing missing data using the approx function? I mean the no leap year has only 365 days. If you for some unknown reasons you want to fill in value for the NAs one "natural" way is to substitute the NAs on February 29th by the mean of the values on February 28th and Marts 1st. I think there is a na.approx function for that in some package (perhaps zoo). Other metods are available in R: google for R + impute. Best Regards Frede Sendt fra Samsung mobil -------- Oprindelig meddelelse -------- Fra: eliza botto Dato:13/06/2014 20.48 (GMT+01:00) Til: r-help@r-project.org Emne: Re: [R] data format setting Thanks dennis, It worked but I had to do some simple modifications to get to the ultimate format. Now I have a list in the following format $A 2004 2005 2006 2007 2008 2009 2010 .. ... ... .. ... $AY 1967 1968 .... 2000........... some columns had 365 rows and some 366. those with 365 rows had their 366 row as NA. Now I want to apply approx. command to interpolate 366 values to 365, but when I apply approx command I gives out something which is with $x and $y, and frankly speaking it messed up everything. Is their a way that i do it neatly without getting the format deteriorated? In any way, thank-you very much indeed. Eliza > Date: Fri, 13 Jun 2014 11:11:37 -0700 > Subject: Re: [R] data format setting > From: djmu...@gmail.com > To: eliza_bo...@hotmail.com > > Hi: > > Maybe something like this: > > library(reshape2) > L <- split(DF, DF$year) > L2 <- llply(L, function(d) dcast(d, month + day ~ year, value.var = > "discharge")) > > Obviously untested, so caveat emptor. The idea is to use the dcast > function to reshape the data from long to wide format within year. > > Dennis > > On Fri, Jun 13, 2014 at 8:55 AM, eliza botto <eliza_bo...@hotmail.com> wrote: > > > > Dear R family, > > I hope you all be doing great. I have a dataset of following format. The > > data file is of the following format. > > > > st year month day discharge > > 1 A 2004 1 1 6.752828 > > 2 A 2004 1 2 7.602053 > > 3 A 2004 1 3 5.583619 > > 4 A 2004 1 4 5.019562 > > 5 A 2004 1 5 4.804489 > > 6 A 2004 1 6 4.363541 > > 7 A 2004 1 7 3.801333 > > 8 A 2004 1 8 3.455991 > > 9 A 2004 1 9 3.402634 > > 10 A 2004 1 10 3.250693 > > ...... > > ...... > > continue > > ...... > > ...... > > st year month day discharge > > 20000 AY 1967 10 3 0.56 > > 20001 AY 1967 10 4 0.56 > > 20002 AY 1967 10 5 0.48 > > 20003 AY 1967 10 6 0.56 > > 20004 AY 1967 10 7 0.48 > > 20005 AY 1967 10 8 0.40 > > 20006 AY 1967 10 9 0.40 > > 20007 AY 1967 10 10 0.56 > > 20008 AY 1967 10 11 0.56 > > 20009 AY 1967 10 12 0.65 > > 20010 AY 1967 10 13 0.85 > > > > you can see that there are five columns. > > The first column has the name of the station. I want to split the data > > w.r.t the names of the stations. Each station has data for certain years. > > for example "A" has data for years from 2004 to 2010 and for "AY" its from > > 1967 to 2000.similarly for other years there is data for different number > > of years. > > I want to make a list of matrices each containing the data for that station > > in the following format > > $A > > 2004 2005 2006 2007 2008 2009 2010 > > .. > > ... > > ... > > .. > > ... > > > > $AY > > > > 1967 1968 .... 2000 > > > > each column should have 365 to 366 values depending on whether there is a > > leap year or not. obviously for non-leap years 366th row should be an NA. > > kindly help me on it. > > Thankyou very much in advance. > > Eliza > > > > > > [[alternative HTML version deleted]] > > > > ______________________________________________ > > R-help@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. [[alternative HTML version deleted]] ______________________________________________ R-help@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. [[alternative HTML version deleted]] ______________________________________________ R-help@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.
