How to loop through all the columns in dataframe

Glad I could help.  You might want to post it back
to R-Help so that
others can see what was done.

On Sun, Mar 16, 2008 at 6:02 PM, Felipe Carrillo
<mazatlanmexico at yahoo.com> wrote:

Jim: I owe you man, this is great,I never thought

that

I could acomplish this task. Now I can estimate
confidence intervals for mydf and I will be
done,,Thanks again Jim..

This should do what you want: (you had td and pd
reversed in your example)

xd <-

c(2.2024,2.4216,1.4672,1.4817,1.4957,1.4431,1.5676)

td <-

pd <-

c(160524,163565,143973,111956,89677,95269,81558)

 mydf<-data.frame(xd,pd,td)
 trans<-t(mydf)
 trans

          [,1]        [,2]        [,3]       

[,4]

    [,5]
[,6]       [,7]
xd 2.20240e+00 2.42160e+00 1.46720e+00

1.48170e+00

1.4957e+00
1.4431e+00 1.5676e+00
pd 1.60524e+05 1.63565e+05 1.43973e+05

1.11956e+05

8.9677e+04
9.5269e+04 8.1558e+04
td 1.70460e-02 1.85040e-02 1.21570e-02

1.22530e-02

1.2348e-02
1.1997e-02 1.2825e-02

 varA<- 0.0000036084
 covAB<- (-0.0000013046)
 varB<- 0.00000052628

# create the sequences to test against
i.seq <- lapply(seq(ncol(trans) - 1),

function(x)

x:(ncol(trans) - 1))

x <- lapply(i.seq, function(.col){

+     # compute the 3 columns of data
+     cbind(xp=varA + trans[1, .col[1]] * covAB

+

trans[1, .col + 1] *
covAB + trans[1, .col[1]] * trans[1, .col + 1] *
varB,
+         pd=trans[2, .col[1]] * trans[2, .col +

1],

+         td=trans[3, .col[1]] * trans[3, .col +

1])

+ })

# rbind for the output
z <- do.call(rbind, x)
# add the covariance
z <- cbind(z, cov=z[, 'xp'] * z[, 'pd'] / z[,

'td'])

z

 Felipe D. Carrillo
 Fishery Biologist
 Department of the Interior
 US Fish & Wildlife Service
 California, USA



    

Be a better friend, newshound, and

-- 
Jim Holtman
Cincinnati, OH
+1 513 646 9390

What is the problem you are trying to solve?