I have the following matrix (let's call it df), for which I would like to create bootstrapped means and 95% confidence intervals for each column, due to the heavily 0 weighted distribution. I would like the mean and CI's to be added to the bottom of the matrix as new rows. This is a small subset of the data, the true data has >600 rows which will make the bootstrapping much more effective.
row.names V183 V184 V185 V186 V187 V188 V189 V190 V191 V192 V193 V194 V195 V196 V197 V198 V199 V200 V201 V202 V203 V204 V205
1 0.07142857 0.07142857 0.07142857 0.07142857 0.07142857 0.07142857 0.07142857 0.07142857 0.07692308 0.07692308 0.07692308 0.07692308 0.07692308 0.07692308 0.07692308 0.07692308 0.07692308 0.07692308 0.07692308 0.07692308 0.07692308 NA NA
2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
3 0.022 0.022 0.022 0.022 0.022 0.022 0.022 0.022 0.022 0.022 0.022 0.022 0.022 0.022 0.022 0.022 0 NA NA NA NA NA NA
4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.07692308 0.07692308 0.07692308 0.07692308 0.07692308 0.07692308 0.07692308 0.07692308
5 0 0 0 0 0.066 0.066 0.066 0.066 0.066 0.066 0.066 0.066 0.066 0.066 0 0 0 0 0 0 0 0 0
6 0.077 0.077 0.077 0.077 0.077 0.077 0.077 0.077 0.077 0.077 0.077 0.077 0.077 0.077 0.077 0.077 0.077 0.077 0.077 0 0 0 0
7 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
8 0.07142857 0.07142857 0.07142857 0.07142857 0.07142857 0.07142857 0.07142857 0.07142857 0.07142857 0.07142857 0.07142857 0 0 0 0 0 0 0 0 0 0 0 0
9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 NA NA NA NA NA NA
10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
11 NA NA NA NA NA NA NA NA NA NA NA NA 0.03225806 0.03225806 0.03225806 0.03225806 0.03225806 0.03225806 0.03225806 0.03225806 0.03225806 0.03225806 0.03225806
12 0 0 0 0 0 0 0 0 0 0 0 0 0 NA NA NA NA NA NA NA NA NA NA
13 0 0 0 0 0 0 0 0 0 NA NA NA NA NA NA NA NA NA NA NA NA NA NA
14 0 0 0.033 0.033 0.033 0.033 0.033 0.033 0.033 0.033 0.033 0.033 0.033 0.033 0.033 0.033 0.033 0 0 0 0 0 0
I have tried this:
boot.mean <- function(df,i){boot.mean <- mean(df[i])}
df["BootMean" ,] <- boot(df, boot.mean, R = 2000)
But it says "undefined columns are selected
So I tried this:
boot.mean <- function(df[1:23],i){boot.mean <- mean(df[i])}
df["BootMean" ,] <- boot(df, boot.mean, R = 2000)
But it says there is a "[" that it doesn't like.
I recently tried this:
n<-length(df)
B<-1000
boot.mean <- function(df,i){boot.mean <- mean(df[,i],na.rm = TRUE)}
df["BootMean" ,] <-for (i in 1:n) {
boot(df[1:14,i],boot.mean,R=B)
}
But I receive a "error in evaluating the argument 'x' in selecting a method for function 'mean': Error in df[, i] : incorrect number of dimensions"
Do I need to use an apply function or something??? Please Help, the brain is hurting over this trivial problem!
*****I've made some progress, but am not all the way yet.
I've been able to get a booted mean for a single row by subsetting it out, but I am unable to incorporate a na.rm=T function into the formula, so I also have to manually remove those. Can anyone suggest a way to add the na.rm fn?
df<-subset(dfboot,F_BS_sub[1:323, 1]>=0)
dfa<-df[,1]
dfb<-subset(dfa,V183>=0)
boot.mean <- function(dfb, d) {
E=dfb[d,]
return(mean(E))}
b = boot(dfb, boot.mean, R=1000)
b
