Specifying number of trials, bootstrap - r

Question

For an assignment, I am applying mixture modeling with the mixtools package on R. When I try to figure out the optimal amount of components with bootstrap. I get the following error
Error in boot.comp(y, x, N = NULL, max.comp = 2, B = 5, sig = 0.05, arbmean = TRUE, :
Number of trials must be specified!
I found out that I have to fill an N: An n-vector of number of trials for the logistic regression type logisregmix. If
NULL, then N is an n-vector of 1s for binary logistic regression.
But, I don't know how to find out what the N is in fact to make my bootstrap working.
Link to my codes:
https://www.kaggle.com/blastchar/telco-customer-churn
My codes:
data <- read.csv("Desktop/WA_Fn-UseC_-Telco-Customer-Churn.csv", stringsAsFactors = FALSE,
na.strings = c("NA", "N/A", "Unknown*", "NULL", ".P"))
data <- droplevels(na.omit(data))
data <- data[c(1:5032),]
testdf <- data[c(5033:7032),]
data <- subset(data, select = -customerID)
set.seed(100)
library(plyr)
library(mixtools)
data$Churn <- revalue(data$Churn, c("Yes"=1, "No"=0))
y <- as.numeric(data$Churn)
x <- model.matrix(Churn ~ . , data = data)
x <- x[, -1] #remove intercept
x <-x[,-c(7, 11, 13, 15, 17, 19, 21)] #multicollinearity
a <- boot.comp(y, x, N = NULL, max.comp = 2, B = 100,
sig = 0.05, arbmean = TRUE, arbvar = TRUE,
mix.type = "logisregmix", hist = TRUE)
Below there is more information about my predictors:
dput(x[1:4,])
structure(c(0, 1, 1, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1,
34, 2, 45, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0,
0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0,
0, 0, 1, 1, 0, 29.85, 56.95, 53.85, 42.3, 29.85, 1889.5, 108.15,
1840.75), .Dim = c(4L, 23L), .Dimnames = list(c("1", "2", "3",
"4"), c("genderMale", "SeniorCitizen", "PartnerYes", "DependentsYes",
"tenure", "PhoneServiceYes", "MultipleLinesYes", "InternetServiceFiber optic",
"InternetServiceNo", "OnlineSecurityYes", "OnlineBackupYes",
"DeviceProtectionYes", "TechSupportYes", "StreamingTVYes", "StreamingMoviesYes",
"ContractOne year", "ContractTwo year", "PaperlessBillingYes",
"PaymentMethodCredit card (automatic)", "PaymentMethodElectronic check",
"PaymentMethodMailed check", "MonthlyCharges", "TotalCharges"
)))
My response variable is binary
I hope you guys can help me out!

Answer 1

Looking in the source code of mixtools::boot.comp, which is scary as it is over 800 lines long and in serious need of refactoring, the offending lines are:
if (mix.type == "logisregmix") {
if (is.null(N))
stop("Number of trials must be specified!")
Despite what the documentation says, N must be specified.
Try to set it to a vector of 1s: N = rep(1, length(y)) or N = rep(1, nrow(x))
In fact, if you look in mixtools::logisregmixEM, the internal function called by boot.comp, you'll see how N is set if NULL:
n <- length(y)
if (is.null(N)) {
N = rep(1, n)
}
Too bad this is never reached if N is NULL since it stops with an error before. This is a bug.