LCA number of parameters & degrees of freedom

Question

I have a series of physicians' claims submissions. I would like to perform cluster analysis as an exploratory tool to find patterns in how physicians bill based on things like Revenue Codes, Procedure Codes, etc. The data are all polytomous, and from my basic understanding, a latent class algorithm is appropriate for this kind of data. I am trying my hand at some of R's cluster packages, & specifically poLCA & mclust for this analysis. I'm getting alerts after running a test model on a sample of the data using poLCA.

> library(poLCA)
> # Example data structure - actual test data has 200 rows:
> df <- structure(list(RevCd = c(274L, 320L, 320L, 450L, 450L, 450L, 
636L, 636L, 636L, 450L, 450L, 450L, 301L, 305L, 450L, 450L, 352L, 
301L, 300L, 636L, 301L, 450L, 636L, 636L, 307L, 450L, 300L, 300L, 
301L, 301L), PlaceofSvc = c(23L, 23L, 23L, 23L, 23L, 23L, 23L, 
23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L, 
23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L), TypOfSvc = c(51L, 
51L, 51L, 51L, 51L, 51L, 51L, 51L, 51L, 51L, 51L, 51L, 51L, 51L, 
51L, 51L, 51L, 51L, 51L, 51L, 51L, 51L, 51L, 51L, 51L, 51L, 51L, 
51L, 51L, 51L), FundType = c(3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 
3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 
3L, 3L, 3L, 3L, 3L, 3L), ProcCd2 = c(1747L, 656L, 656L, 1375L, 
1376L, 1439L, 1623L, 1645L, 1662L, 176L, 1374L, 1376L, 958L, 
1032L, 1368L, 1374L, 707L, 960L, 347L, 1662L, 859L, 1375L, 1654L, 
1783L, 882L, 1440L, 332L, 332L, 946L, 946L)), .Names = c("RevCd", 
"PlaceofSvc", "TypOfSvc", "FundType", "ProcCd2"), row.names = c(1137L, 
1138L, 1139L, 1140L, 1141L, 1142L, 1143L, 1144L, 1145L, 1146L, 
1147L, 1945L, 1946L, 1947L, 1948L, 1949L, 1950L, 1951L, 1952L, 
1953L, 1954L, 1955L, 1956L, 1957L, 1958L, 1959L, 2265L, 2266L, 
2267L, 2268L), class = "data.frame")

> clust <- poLCA(cbind(RevCd, PlaceofSvc, TypOfSvc, FundType, ProcCd2)~1, df, nclass = 3)

========================================================= 
Fit for 3 latent classes: 
========================================================= 
number of observations: 200 
number of estimated parameters: 7769 
residual degrees of freedom: -7569 
maximum log-likelihood: -1060.778 

AIC(3): 17659.56
BIC(3): 43284.18
G^2(3): 559.9219 (Likelihood ratio/deviance statistic) 
X^2(3): 33852.85 (Chi-square goodness of fit) 

ALERT: number of parameters estimated ( 7769 ) exceeds number of observations ( 200 ) 

ALERT: negative degrees of freedom; respecify model 

My novice assumption is that I need to run a greater number of iterations before I can get results that are robust? e.g. "...it is essential to run poLCA multiple times until you can be reasonably certain that you have found the parameter estimates that produce the global maximum likelihood solution." (http://www.sscnet.ucla.edu/polisci/faculty/lewis/pdf/poLCA-JSS-final.pdf). Alternatively, perhaps certain variables, particularly CPT & Revenue Codes, have too many unique values, and that I need to aggregate these variables into higher level categories to reduce the number of parameters?

When I run the model using package mclust, which optimizes the model based on BIC, I don't get any such alert.

> library(mclust)
> clustBIC <- mclustBIC(df)
> summary(clustBIC, data = df)

classification table:
      1   2 
     141  59 

 best BIC values:
        VEV,2     VEV,3     EEV,3 
      -4562.286 -4706.190 -5655.783

If anyone can shed a bit of light on the above alerts, it would be much appreciated. I was also planning on using the script found in the poLCA documentation to run multiple iterations of the model until the log-likelihood is maximized. However it's computationally intensive and I'm afraid the process will crash before I have a chance to post this. Sorry in advance if I've missed something obvious here; I'm new to cluster analysis.

Answer 1

polca and mclust both performs Model-based cluster analysis, based on finite mixture models. However, polca is designed for Latent Class Analysis (LCA) which is the name for a particular class of mixture models suitable for categorical (polytomous) data. On the converse, mclust estimates Gaussian mixtures, so is suitable for quantitative variables.

You should choose between the two classes of models by analyzing the nature and structure of your variables. Note that with LCA you are considering the variables as qualitative, that is, the information about the ordering of the modalities is ignored.

As regards to poLCA, you have too many unique values in each variable for the model to be identifiable. The number of independent parameters is related to the number of modalities (what you called unique values) of each variable and must be lower than the number of distinct configurations of the variables (in your case distinct observed 5-ples of outcomes among the units, which is $\leq 200$). In particular, if $m_a$, $m_b$, $m_c$ are the numbers of modalities for a 3-variables models with $k$ Latent Classes, then the number of independent parameters is: $$ (k-1)+ k\cdot[(m_a-1)+(m_b-1)+(m_c-1)] $$ So, yes: if you want to use LCA, you need to aggregate the modalities in order to reduce the number of parameters.

Btw, to run poLCA multiple times, you can simply use the nrep option.

Answer 2

As has been noted, poLCA only handles categorical data. So what does that mean for how it processes your data? From the help documentation for poLCA:

...Manifest variables must contain only integer values, and must be coded with consecutive values from 1 to the maximum number of outcomes for each variable.

That means that for each of your variables, poLCA will assume each value is a unique category, and that there are as many possible categories (outcomes) as the highest value in the variable. Taking a look at the data you provided, that would imply that RevCd is a categorical variable with more than 600 unique levels! There appear to be similar problems with ProcCd2, and possibly the other three variables as well (it's hard to be certain without knowing more about the structure of your data set).

mclust, on the other hand, is designed to work only with continuous data. That means that when you give it a categorical variable, it just processes it as though it were continuous (which is almost certainly not what you want). Assume your variable FundType is categorical with three levels 1, 2, and 3 corresponding to three different types of funds. If you process that as though it were continuous, then you're not estimating anything to do with three different funds, you're estimating some made up underlying continuous variable (let's call it "fundness"), which increases linearly from fund 1 to fund 2, and fund 2 to fund 3. Except in some rare cases with ordinal categorical variables, this is nonsense.

My guess is that you want to use a combination of both continuous and categorical variables to determine latent classes. There's no mathematical reason you can't do this, but neither poLCA nor mclust will run the model you need. depmixs4 can handle both categorical and continuous manifest variables, so you may want to start there. Also see this post with a related question: Latent class model with both continuous and categorical indicators in R