[R] overwritten plots in pdf file

i am having a problem in saving plots in pdf. i have this code below and it only shows me the last plot. i tried keeping my devices open by removing dev.off() from the code but the pdf file won't open for (i in 1:n) { . . . pdf("D:/research/plot.pdf") plot(mon, mu, type ='o') dev.off() } my for

[R] neural network arguments

hi everyone! my inquiry with neural network is rather basic. i am just learning neural network, particularly the VR bundle. i read the documentations for the said bundle but still is struggling on understanding some arguments - x is the matrix or data frame of x values for example does this m

[R] acf for a univariate time series in a data frame

hi everyone! i want to check the autocorrelation function for a univariate time series (streamflow) in a data frame as below: < DF <- read.table("D:/file path") < DF year jan feb mar apr .. dec 1966 0.504 0.406 0.740 0.241 0.429 1967 0.683 0.529

Re: [R] acf for a univariate time series in a data frame

*additional: the lags i am expecting is in months.. since i am trying to predict monthly streamflow. thanks again -- View this message in context: http://www.nabble.com/acf-for-a-univariate-time-series-in-a-data-frame-tp25799751p25799784.html Sent from the R help mailing list archive at Nabble.

[R] simulating AR() using a

good day everyone! i have a time series (andong.ts) and fitted and AR() model using the following code andong.ts <- ts(read.table("D:/.../andong.csv", header = TRUE), start = c(1966,1), frequency = 1) ar(andong.ts) Call: ar(x = andong) Coefficients: 1 2 3 0.3117 0.0607 0.

[R] help in simulating AR models

good day everyone! i have a time series (andong.ts) and fitted and AR() model using the following code andong.ts <- ts(read.table("D:/.../andong.csv", header = TRUE), start = c(1966,1), frequency = 1) ar(andong.ts) Call: ar(x = andong) Coefficients: 1 2

[R] how to create normal qqplot with the 95% confidence interval

hi everyone! season's greetings! is there any way that i can create a normal qqplot showing, aside from the qqline, the 95% confidence limits? thank you very much.. happy holidays! -- View this message in context: http://n4.nabble.com/how-to-create-normal-qqplot-with-the-95-confidence-interva

Re: [R] how to create normal qqplot with the 95% confidence interval

thank you very much! this is just what my professor was asking for. Walmes Zeviani wrote: > > Rubio, > > > Look at library(fBasics) the function qqnormPlot(). Below an example: > > qqnormPlot(rnorm(100)) > > Best's > > Walmes Zeviani, Brasil. >

[R] help in identifying the argument "formula" in the package nnet

ot; for my neural network.. I am aware that my model should be Q(t) ~ Q(t-1) + Q(t-2) + Q(t-3). But I can't seem to figure out how to express this in a formula that R would recognize. Any help, suggestions and comments to enlighten me will be truly appreciated. CJ Rubio -- View this message in

[R] how to express time series linear model Q(t) ~ Q(t-1)+.. Q(t-n) as a formula

For example I have a time series Q(t) ~ Q(t-1) + Q(t-2) + Q(t-3) meaning that my current value is dependent to the 3 previous values. Can anybody help me express this in a formula that I can use for my neural network model (I am planning to use packages "nnet" and "MASS") -- View this message

[R] annual maximum value

hi everyone! hope you can help me here. i am a new R user. what i am trying to do is to find the maximum annual discharge from a daily record. i have a data.frame which includes date and the discharge. somewhat like this.. 10/1/1989 2410 10/2/1989 2460 10/3/1989 2890 ... ...

Re: [R] annual maximum value

m[,4],as.Date(as.character(DF[,3]), "%m/%d/%Y")) > x <- aggregate(z, cut(time(z),"y"), max) i got the maximum values right using coredata(), my question now is, how can i call for the complete date (-mm-dd) and the year () when the maximum observation for that

[R] problem with dates in zoo package

i have the following code - assimilating the maximum annual discharge each year ffrom a daily discharge record from year 1989-2005. m <- read.table("D:/documents/5 stations/01014000.csv", sep =",") z <- zoo(m[,4],as.Date(as.character(m[,3]), "%m/%d/%Y")) x <- aggregate(z, floor(as.numeric(as.yea

[R] rbind: number of columns of result is not a multiple of vector length (arg 1)

i have the following constructed and running very well,, thanks to Gabor Grothendieck for his help. >data.info <- c("station.id", "year", "date", "max.discharge") > > for(i in 1:num.files) { + station.id <- substring(data[i], 1,8) + DF <- read.table(data[i], sep=",", blank.lines.skip = TRUE) + z

Re: [R] rbind: number of columns of result is not a multiple of vector length (arg 1)

ta.info, cbind(station.id, year, date, > max.discharge)) > > On Tue, Feb 17, 2009 at 9:26 PM, CJ Rubio wrote: >> >> i have the following constructed and running very well,, thanks to Gabor >> Grothendieck for his help. >> >>>data.info <- c("

[R] basic inquiry regarding write.csv

i have a loop which looks likes this: > data.info <- rbind(data.info, cbind(station.id, year, date, > max.discharge)) + y <- split(data.info, data.info[station.id]) + for (i in names(y)) {write.csv(y[[i]], file=paste(i, ".csv", sep=","))} i am wondering, where the file (which i am ab

Re: [R] basic inquiry regarding write.csv

; If the "file" is a relative path, then it should be in the working > directory. Say, the working directory is E:/my_work_directory (of > course, you can get it by getwd()), and you export a data frame "a" to > csv by: > write.csv(a, file="a.csv"), then th

Re: [R] basic inquiry regarding write.csv

tion you gave me.. and there were no errors when i run it. hope this would further clarify my question. So far, I cannot see any mistake, though the sep="" will be more elegant that sep=",". Are you sure your working directory is "E:/my_work_directory"? and is there any