I am trying to write a code which evaluates 3 columns of my dataframe rowwise and iteratively, and produces a new column with the minimum value found in the evaluation. Here's what my data looks like:
mydata
Jan Feb Mar Apr May Jun Jul Aug
1 .25 .50 .50 .25 .75 .50 .25 .25
2 .75 .25 .75 .50 .25 .50 .50 .50
3 .50 .50 .50 .50 .25 .25 .75 .25
4 .25 .75 .75 .50 .75 .75 .75 .75
And this is what i want to produce:
mydata
Jan Feb Mar Apr May Jun Jul Aug -Mar ... -Aug
1 .25 .50 .50 .25 .75 .50 .25 .25 .25 ... .25
2 .75 .25 .75 .50 .25 .50 .50 .50 .25 ... .50
3 .50 .50 .50 .50 .25 .25 .75 .25 .50 ... .25
4 .25 .75 .75 .50 .75 .75 .75 .75 .25 ... .75
With -Mar showing the minimum valuess of (Jan:Mar), -Apr showing the minimum value of (Feb:Apr), and so on.
I tried the following:
> for (i in 3:8)
{
Newcolname <- as.name(paste("-",i,sep=""))
mydata$Newcolname<- pmin(mydata[,i-2],mydata[,i-1],mydata[,i])
}
But this only gives my 1 new column, containing the minimum values for (jun:aug), being the last 3 columns of mydata. What I want instead is 5 new columns as depicted above, but I cannot figure out how to do this. Anybody?
Please let me know if i'm not clear. Thanks in advance!
You can try this:
for (i in 3:8) {
Newcolname <- paste0("-", names(mydata)[i])
mydata[[Newcolname]] <- pmin(mydata[,i-2],mydata[,i-1],mydata[,i])
}
where names(mydata)[i] is the i'th name from mydata, e.g. names(mydata)[3] is Mar. To create a new column you have to put the variable in [[ ]]. The way you had it, it created a new column named Newcolname.