Resolving warning to increase random starts using Mplus automation

[eshuman1650]

Hi,

I am trying to run model 3 with three classes using the Mplus automation, using this code:

d4b %>%
   estimate_profiles(3,
                     package = "MplusAutomation",
                     variances = c("equal"),
                     covariances = c("equal"),
                     keepfiles = T)

When I run it I get an warning: "One or more analyses resulted in warnings! Examine these analyses carefully: model_3_class_3". I examined the Mplus output (see attached) and the warning was because of the warning message: "WARNING: THE BEST LOGLIKELIHOOD VALUE WAS NOT REPLICATED. THE SOLUTION MAY NOT BE TRUSTWORTHY DUE TO LOCAL MAXIMA. INCREASE THE NUMBER OF RANDOM STARTS." So I increased the number of random starts using the ANALYSIS = "starts = 4000 1000;" in the estimate_profiles function in R:

d4b %>%
   estimate_profiles(3,
                     package = "MplusAutomation",
                     ANALYSIS = "starts = 4000 1000;",
                     variances = c("equal"),
                     covariances = c("equal"),
                     keepfiles = T)

When I run this I still get the same warning in R, but when I examine the Mplus output the only warning is: "WARNING in MODEL command: All variables are uncorrelated with all other variables within class. Check that this is what is intended." This warning always appears, even in other models where I don't get the warning in R. So I am trying to understand why I still get the warning in R even after I increase the number of random starts.

Thanks, Eric

model_3_class_3_reg.txt model_3_class_3_4000_1000.txt