尽管没有缺失值，但仍获得NA / NaN错误消息（使用R：tidyLPA）

Question

I'm using tidyLPA in R. I followed the example given here . The example code they give is the following:

library(tidyLPA)
library(dplyr)
library(mclust)    
data(pisaUSA15)

pisaUSA15[1:100, ] %>%
  select(broad_interest, enjoyment, self_efficacy) %>%
  single_imputation() %>%
  estimate_profiles(3)

This runs fine and gives me the following results:

tidyLPA analysis using mclust: 

 Model Classes AIC     BIC     Entropy prob_min prob_max n_min n_max BLRT_p
 1     3       630.957 667.430 0.885   0.903    0.973    0.130 0.680 0.010

However, when I run it with my own data using the following code:

data[1:100, ] %>%
  select(V1, V2, V3, V4) %>%
  single_imputation() %>%
  estimate_profiles(3)

I get the following error message:

Error in mdpwst[i]:mdpwfin[i] : NA/NaN argument

I'm confused as to why I'm getting an NA/NaN message when I deleted all observations with missing values with:

data = data[complete.cases(data),]

This has never happened to me with any other package and sifting through similar questions here on stackoverflow provided no answers. I'm not sure how to address this and just looking at the datasets show no immediately identifiable reasons as to why one dataset is working and one is not.

Answer 1

I'm answering my own question here, The issue came from the single_imputation() %>% command. As stated here on Github , the command is used when there are missing values. Since I removed all my missing values, the command is not needed and is actually causing an error!

Once I removed it to have:

data[1:100, ] %>%
  select(V1, V2, V3, V4) %>%
  estimate_profiles(3)

The code ran fine!