AC Del Re
2011-Nov-03 15:54 UTC
[R] Take variables in data.frame and create list of matrices
Hi, I have this sample data below and would like to create a list of matricies. setseed(1254) id <- c(1,1,1,1 ,2,2,2) o <- as.factor(c(1:4, 1, 3, 4)) r <- rep(.5, 7) v <- rnorm(7) s <- rnorm(7) dat <-data.frame(id, o, r, v, s) dat #> dat # id o r v s # 1 1 0.5 0.7024631 2.0813672 # 1 2 0.5 -0.5541955 0.1095156 # 1 3 0.5 -1.0418167 0.4164930 # 1 4 0.5 1.9973868 -2.0402058 # 2 1 0.5 1.3957243 -0.2548652 # 2 3 0.5 -0.6571325 0.6946570 # 2 4 0.5 1.3998202 0.4601017 Is there an easy way (or convenient function) to convert this into a list where each unique 'id' has its own 4x4 matrix (based on the maximum number of factor levels in 'o')in the list? Each matrix would have v in the diagonal and r*s*s in the off-diagonal (e.g., .5*2.0814*.1095 = 0.1139), where r is constant (.5) and s varies based on the wthin 'id' factor level of 'o'. The output would look something like this (except values in all cells): [[id 1]] 'o'-levels [,1] [,2] [,3] [,4] [1,] .702 .1139 r*s1*s3 ... [2,] .1139 .702 r*s2*s3 ... [3,] r*s3*s1 ... . 702 [4,] .702 [[id 2]] 'o'-levels [,1] [,2] [,3] [,4] [1,] .702 NA r*s1*s3 ... [2,] NA .702 NA ... [3,] r*s3*s1 NA .702 [4,] ... .702 any assistance is much appreciated. Thank you, AC [[alternative HTML version deleted]]
jose Bartolomei
2011-Nov-03 16:20 UTC
[R] Take variables in data.frame and create list of matrices
If i understood correctly this may help you > dlply(dat, .(id)) $`1` id o r v s 1 1 1 0.5 -1.7131858 -2.0548666 2 1 2 0.5 0.6582979 0.9688939 3 1 3 0.5 -0.6861830 2.4213072 4 1 4 0.5 1.4134171 -0.4540244$`2` id o r v s 5 2 1 0.5 0.2771839 1.0688474 6 2 3 0.5 -0.7543462 -1.0371494 7 2 4 0.5 -1.9810018 -0.6742232 > Date: Thu, 3 Nov 2011 08:54:52 -0700> From: delre@wisc.edu > To: r-help@r-project.org > Subject: [R] Take variables in data.frame and create list of matrices > > Hi, > > I have this sample data below and would like to create a list of matricies. > > setseed(1254) > id <- c(1,1,1,1 ,2,2,2) > o <- as.factor(c(1:4, 1, 3, 4)) > r <- rep(.5, 7) > v <- rnorm(7) > s <- rnorm(7) > > dat <-data.frame(id, o, r, v, s) > dat > > #> dat > # id o r v s > # 1 1 0.5 0.7024631 2.0813672 > # 1 2 0.5 -0.5541955 0.1095156 > # 1 3 0.5 -1.0418167 0.4164930 > # 1 4 0.5 1.9973868 -2.0402058 > # 2 1 0.5 1.3957243 -0.2548652 > # 2 3 0.5 -0.6571325 0.6946570 > # 2 4 0.5 1.3998202 0.4601017 > > Is there an easy way (or convenient function) to convert this into a list > where each unique 'id' has > its own 4x4 matrix (based on the maximum number of factor levels in 'o')in > the list? > Each matrix would have v in the diagonal and r*s*s in the off-diagonal > (e.g., > .5*2.0814*.1095 = 0.1139), where > r is constant (.5) and s varies based on the wthin 'id' factor level of > 'o'. > The output would look something like this (except values in all cells): > > [[id 1]] > > 'o'-levels [,1] [,2] [,3] [,4] > > [1,] .702 .1139 r*s1*s3 ... > > [2,] .1139 .702 r*s2*s3 ... > > [3,] r*s3*s1 ... . 702 > > [4,] .702 > > > [[id 2]] > > 'o'-levels [,1] [,2] [,3] [,4] > > [1,] .702 NA r*s1*s3 ... > > [2,] NA .702 NA ... > > [3,] r*s3*s1 NA .702 > > [4,] ... .702 > > > any assistance is much appreciated. > > Thank you, > > AC > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.[[alternative HTML version deleted]]
jose Bartolomei
2011-Nov-03 16:25 UTC
[R] Take variables in data.frame and create list of matrices
I missed; using the Hadley Wickham plyr package >library (plyr)> dlply(dat, .(id)) $`1` id o r v s 1 1 1 0.5 -1.7131858 -2.0548666 2 1 2 0.5 0.6582979 0.9688939 3 1 3 0.5 -0.6861830 2.4213072 4 1 4 0.5 1.4134171 -0.4540244$`2` id o r v s 5 2 1 0.5 0.2771839 1.0688474 6 2 3 0.5 -0.7543462 -1.0371494 7 2 4 0.5 -1.9810018 -0.6742232 > Date: Thu, 3 Nov 2011 08:54:52 -0700> From: delre@wisc.edu > To: r-help@r-project.org > Subject: [R] Take variables in data.frame and create list of matrices > > Hi, > > I have this sample data below and would like to create a list of matricies. > > setseed(1254) > id <- c(1,1,1,1 ,2,2,2) > o <- as.factor(c(1:4, 1, 3, 4)) > r <- rep(.5, 7) > v <- rnorm(7) > s <- rnorm(7) > > dat <-data.frame(id, o, r, v, s) > dat > > #> dat > # id o r v s > # 1 1 0.5 0.7024631 2.0813672 > # 1 2 0.5 -0.5541955 0.1095156 > # 1 3 0.5 -1.0418167 0.4164930 > # 1 4 0.5 1.9973868 -2.0402058 > # 2 1 0.5 1.3957243 -0.2548652 > # 2 3 0.5 -0.6571325 0.6946570 > # 2 4 0.5 1.3998202 0.4601017 > > Is there an easy way (or convenient function) to convert this into a list > where each unique 'id' has > its own 4x4 matrix (based on the maximum number of factor levels in 'o')in > the list? > Each matrix would have v in the diagonal and r*s*s in the off-diagonal > (e.g., > .5*2.0814*.1095 = 0.1139), where > r is constant (.5) and s varies based on the wthin 'id' factor level of > 'o'. > The output would look something like this (except values in all cells): > > [[id 1]] > > 'o'-levels [,1] [,2] [,3] [,4] > > [1,] .702 .1139 r*s1*s3 ... > > [2,] .1139 .702 r*s2*s3 ... > > [3,] r*s3*s1 ... . 702 > > [4,] .702 > > > [[id 2]] > > 'o'-levels [,1] [,2] [,3] [,4] > > [1,] .702 NA r*s1*s3 ... > > [2,] NA .702 NA ... > > [3,] r*s3*s1 NA .702 > > [4,] ... .702 > > > any assistance is much appreciated. > > Thank you, > > AC > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.[[alternative HTML version deleted]]