Question

Anyone use R for data manipulation?

I'm trying to get a subset and struggling with my code...

(basically I'm trying to select stuff based on different values in different columns, but I just don't know how to word it).

Answer 1

You'd likely be better off asking this in the CompSci section of the site. And it'd be good to include some source code and data.

Answer 2

R is good for data manipulation?

Answer 3

Yeah I suspected that, I just figured I'd wait to see if anyone used R. Programmers don't really use R do they? And yepp!

Answer 4

I have never used R and nor I am looking forward to using it ... btw what about sql or my sql?

Answer 5

Haven't used sql/my sql before. Nah don't worry R is VERY useful does take some learning but way better than excel once you learn it

Answer 6

use nested if statements that cycle through your data and break if a requirement isn't met?

Answer 7

Thanks for your help, I think I will have to come back to this later when my brain is working a little better, I've been trying to type a description of what I am trying to do and not getting anywhere!

Answer 8

while(!eof) //while you haven't reached the end of the file (eof)
if (mindate<x<maxdate) {
next if
} else { break;}
}
Something like that, I'm not at all familiar with R, I'm more into OOP and functional languages. But, I was meant to study it at one point.

As I said give some code and a description of what you want, and I'll try give some pointers.

Answer 9

Hmmm.. What if I show you the way I was doing it in excel, would that make any sense to you? (p.s. I definitely not from a programing background.. to tell you the thruth I've never used a loop before)

Answer 10

oh dear I can't even spell today!

Anyway if you look to the far right of the sheet (which I cut alot of data out of so you could see better) you will see the excel functions I've used - basically just to get a percentage of each landuse in the "LCT_1990" column out of each "FID_buffer931" code (which is each site - first column). My "FID_buffer931" gets down to 19, so I don't want to keep doing it manually in excel like I am doing.

I was originally trying to stick "q<-subset(B,"LCT.1990=bare" & "FID.poly9093=1")" in R just to select one landuse in one site (yes I can hear you cringe at that awful - and wrong - piece of code).

Answer 11

hmmm and You're sure that R is the language you'd like to use?

Sorry I might be a bit slow at this, your variable names are not terribly stranger friendly. What are you having difficulty with?
Is it where you are using sumif in excel?

Answer 12

Hmm yeah, mainly because its the only other one I remotely know how to use, and I know it can be used for that type of thing.. I'm just missing the how part :P

Yeah, they are pretty bad.. that's why I was reluctant to go on lol. Anyway basically "LCT_1990" = is landuse and "FID_buffer931" = site.

You see I know how to do it in excel - like you can see - but to do the whole dataset (which is much longer) I would have to keep pasting those excel functions under each block of "sites" or "FID_buffer931" numbers (separated by the colours). I suppose I was trying to be lazy :P and I suspect I should probably go to an R forum hehe.

Answer 13

Anyway please don't waste time on it! I'm going to bed :)

Answer 14

In java or c++ once you'd read these in, I would make them in to objects with, then pass them to functions which would handle the calculations.

I'm sorry I was expecting a bit of debugging rather than thinking the whole thing out.