I have a data frame that I am running a Monte Carlo simulation on, using for loops, to generate a simulated distribution. As I am testing the simulation code, I am just accessing the first observation in the data frame:
Male.MC <-c()
for (j in 1:100){
for (i in 1:1) {
# u2 <- Male.DistF$Male.stddev_u2[i] * rnorm(1, mean = 0, sd = 1)
u2 <- Male.DistF$RndmEffct[i] * rnorm(1, mean = 0, sd = 1)
mc_bca <- Male.DistF$lmefits[i] + u2
temp <- Lambda.Value*mc_bca+1
ginv_a <- temp^(1/Lambda.Value)
d2ginv_a <- max(0,(1-Lambda.Value)*temp^(1/Lambda.Value-2))
mc_amount <- ginv_a + d2ginv_a * Male.DistF$Male.var[i]^2 / 2
z <- c(RespondentID <- Male.DistF$RespondentID[i],
Male.DistF$AgeFactor[i], Male.DistF$SampleWeight[i],
Male.DistF$Male.var[i], Male.DistF$lmefits[i], u2, mc_amount)
Male.MC <- as.data.frame(rbind(Male.MC,z))
}
}
colnames(Male.MC) <- c("RespondentID", "AgeFactor",
"SampleWeight", "VarByAge",
"lmefits", "u2", "mc_amount")
The code works beautifully except that Male.DistF$RespondentID is a factor and I don’t get the factor level output, but instead get the factor index, in this case I get 1 as the RespondentIDs are in ascending order in the Male.DistF data frame. I have the same problem with AgeFactor, where I get the index rather than the factor level.
head(Male.MC)
RespondentID AgeFactor SampleWeight VarByAge lmefits u2 mc_amount
z 1 3 0.4952835 0.4189871 15.22634 0.2334501 11582.681
2 1 3 0.4952835 0.4189871 15.22634 0.3205741 11984.220
3 1 3 0.4952835 0.4189871 15.22634 -0.5674165 8420.678
4 1 3 0.4952835 0.4189871 15.22634 -0.5426489 8505.421
5 1 3 0.4952835 0.4189871 15.22634 0.4878695 12790.565
6 1 3 0.4952835 0.4189871 15.22634 0.1556925 11234.583
How can I make the `Male.MC1 data frame contain the factor levels for those two variables? I have tried:
z <- c(RespondentID <- as.character(Male.DistF$RespondentID[i]),
Male.DistF$AgeFactor[i], Male.DistF$SampleWeight[i],
Male.DistF$Male.var[i], Male.DistF$lmefits[i], u2, mc_amount)
and
z <- c((as.character(Male.DistF$RespondentID[i])),
Male.DistF$AgeFactor[i], Male.DistF$SampleWeight[i],
Male.DistF$Male.var[i], Male.DistF$lmefits[i], u2, mc_amount)
to fix the RespondentID output, but I am doing something wrong with that syntax and it’s trying to convert all the output to factors:
There were 50 or more warnings (use warnings() to see the first 50)
str(Male.MC)
'data.frame': 100 obs. of 7 variables:
$ RespondentID: Factor w/ 1 level "100020": 1 1 1 1 1 1 1 1 1 1 ...
..- attr(*, "names")= chr "z" "" "" "" ...
$ AgeFactor : Factor w/ 1 level "3": 1 1 1 1 1 1 1 1 1 1 ...
..- attr(*, "names")= chr "z" "" "" "" ...
$ SampleWeight: Factor w/ 1 level "0.495283471": 1 1 1 1 1 1 1 1 1 1 ...
..- attr(*, "names")= chr "z" "" "" "" ...
$ VarByAge : Factor w/ 1 level "0.418987052181831": 1 1 1 1 1 1 1 1 1 1 ...
..- attr(*, "names")= chr "z" "" "" "" ...
$ lmefits : Factor w/ 1 level "15.2263403968895": 1 1 1 1 1 1 1 1 1 1 ...
..- attr(*, "names")= chr "z" "" "" "" ...
$ u2 : Factor w/ 1 level "-0.100954008424162": 1 NA NA NA NA NA NA NA NA NA ...
..- attr(*, "names")= chr "z" "" "" "" ...
$ mc_amount : Factor w/ 1 level "10151.4582133747": 1 NA NA NA NA NA NA NA NA NA ...
..- attr(*, "names")= chr "z" "" "" "" ...
For testing, here is the first couple of rows of the input data frame Male.DistF:
AgeFactor RespondentID SampleWeight IntakeAmt RndmEffct NutrientID Gender Age BodyWeight IntakeDay BoxCoxXY lmefits lmeres TotWts GrpWts NumSubjects TotSubjects Male.var
1725 9to13 100020 0.4952835 12145.852 0.30288536 267 1 12 51.6 Day1Intake 15.61196 15.22634 0.27138449 2291.827 763.0604 525 2249 0.4189871
203 14to18 100419 0.3632839 9591.953 0.02703093 267 1 14 46.3 Day1Intake 15.01444 15.31373 -0.18039624 2291.827 472.3106 561 2249 0.3365423
Lambda.Value is 0.1.
The information on Male.DistF is:
str(Male.DistF)
'data.frame': 2249 obs. of 18 variables:
$ AgeFactor : Ord.factor w/ 4 levels "1to3"<"4to8"<..: 3 4 3 4 2 2 3 1 1 3 ...
$ RespondentID: Factor w/ 2249 levels "100020","100419",..: 1 2 3 4 5 6 7 8 9 10 ...
$ SampleWeight: num 0.495 0.363 0.495 1.326 2.12 ...
$ IntakeAmt : num 12146 9592 7839 11113 7150 ...
$ RndmEffct : num 0.3029 0.027 0.0772 0.4667 -0.1593 ...
$ NutrientID : int 267 267 267 267 267 267 267 267 267 267 ...
$ Gender : int 1 1 1 1 1 1 1 1 1 1 ...
$ Age : int 12 14 11 15 6 5 10 2 2 9 ...
$ BodyWeight : num 51.6 46.3 46.1 63.2 28.4 18 38.2 14.4 14.6 32.1 ...
$ IntakeDay : Factor w/ 2 levels "Day1Intake","Day2Intake": 1 1 1 1 1 1 1 1 1 1 ...
$ BoxCoxXY : num 15.6 15 14.5 15.4 14.3 ...
$ lmefits : num 15.2 15.3 15 15.8 14.3 ...
$ lmeres : num 0.271 -0.18 -0.342 -0.424 -0.053 ...
$ TotWts : num 2292 2292 2292 2292 2292 ...
$ GrpWts : num 763 472 763 472 779 ...
$ NumSubjects : int 525 561 525 561 613 613 525 550 550 525 ...
$ TotSubjects : int 2249 2249 2249 2249 2249 2249 2249 2249 2249 2249 ...
$ Male.var : num 0.419 0.337 0.419 0.337 0.267 ...
As you can see from my Male.DistF data, for the 100 replicates for the first observation, in the Male.MC data frame I would like 100020 as the RespondentID (and not 1) and 9to13 as the AgeFactor (and not 3). Where have I gone wrong with my output instructions and how do I fix this? In particular, I’m not following why my attempts to use as.character went so badly astray as to affect the entire output. As an aside I would also welcome suggestions to speed up the loops. All I am doing is constructing 100 sets of values for each observation in my Male.DistF data frame.
You could try to replace the line
which creates the new row as a vector,
i.e., forces all elements to have the same type,
with a 1-row data.frame, to keep the type of the columns.