I'd like to subset the most recent 20 years of data from a data frame, in this case "carnation." I thought the following would do this, but it returns only the first row for each year. temp <- unique
Thank you to Jean Adams, Susan Rasch, Petr Pikal, Bob Garrett, Bert Gunter, Nick Ellis, and Julian Taylor for providing solutions to my problem re subsetting. All replied to the list so I'll not summ
my mistake: Nick Ellis, and Julian Taylor were responding on Wednesday to a different subsetting question. Todd Thank you to Jean Adams, Susan Rasch, Petr Pikal, Bob Garrett, Bert Gunter, Nick Ellis,
The GUI has a stack columns command with the option to stack ALL columns in a dataframe or matrix. Can someone please tell me how to do this from the command line (Splus 6.0, Win98)? Thanks for any h
Sincere thanks to Nick Ellis for his blisteringly fast response to my query. His "reshape" function (pasted below) does exactly what I want. Todd Hatfield reshape(air,,1:4) where reshape <- function(
I have a dataframe with two columns: a column with dates (in daily steps for 15 years), and a column with corresponding reservoir levels. Rather than plotting the data as a continuous time series, I'
Sincere thanks to: Anne York, James Holtman, Renaud Lancelot, Deb Carlson, and Patrick Connolly for helping me with my question regarding overlaying time series data in a plot. There were a number of
After seeing Gerald Jean's speed improvements yesterday, I've been suffering from fits of jealousy. I'm running some straightforward Monte Carlo simulations, but they are painfully slow because I ca
I'm told my first post was in html, for which I apologize. Here it is again in plain text. Thanks! Greetings. After seeing Gerald Jean's speed improvements yesterday, I've been suffering from fits of
Thank you very much to Bill Dunlap, Tim Hesterberg, and Jingshan Zhang for their help with my looping problem. Bill showed me several ways to make the loop more efficient, and Jingshan pointed out th
I am using S+ 2000 R2 on Win98. I have 1500 dome-shaped curves from a variety of simulations and empirical measurements (not mine). For each curve I digitized the peak and four points on each side of
Yesterday I posted a question with a subject header "fitting a normal to a few digitized points?" and have been asked by Bill Venables for some more information crucial to addressing the problem. In
A very BIG thank you to Bill Venables for his kind help and encouragement. This is a great way to start my week! Below I post the original question and Bill's reply. All the best, Todd Hatfield == Ye