I have a vector of numeric values. My hypothesis is that this vector is a mixture drawn from two Gaussian distributions (ie k = 2). However, it is possible that there is only one Gaussian underlying my data (k = 1). I am attempting to answer this question in a data-driven manner but do not know the best method?

My thought was to compare the two methods by calculating the BIC or AIC for each, and then performing a log-likelihood test.

  1. Should I include k as one of the parameters being estimated when I calculate BIC (ie {mu1, sd1, mu2, sd2, k} vs {mu1, sd1, k} for the 2-component and 1-component models respectively)

  2. I'm using the mixtools package in R and the normalmixEM() function does not seem to allow fitting a 1-component gaussian (ie if I use k = 1 I get an error arbmean and arbvar cannot both be FALSE)

  3. If using a LR with AIC/BIC is not appropriate, is there a more appropriate solution to this problem?

Edit: I found a somewhat illuminating example here. This approach uses the mclust package to fit a 1 vs 2 component gaussian mixture and use the model log-likelihood to perform a likelihood ratio test.

  • $\begingroup$ This paper and this paper should be helpful, they also have a R package: MixtureInf $\endgroup$
    – Francis
    Dec 20, 2017 at 13:26
  • $\begingroup$ Those are indeed helpful, many thanks! I will take a look at their implementation in the package. $\endgroup$
    – Brandon
    Dec 20, 2017 at 17:02
  • $\begingroup$ @Brandon, were you able to figure out how to implement these packages to find the ideal number of distributions to use? $\endgroup$
    – Max F
    Jul 25, 2022 at 19:34
  • 1
    $\begingroup$ @MaxF It's been a while and no longer involved in that lab so I don't recall exactly how I resolved it. I believe I ended up using the mclust approach from the link I provided in the original post edit. $\endgroup$
    – Brandon
    Jul 25, 2022 at 21:23

1 Answer 1


An alternative strategy is to test for Normality. If your data comes from a single Gaussian, you should fail to reject the null hypothesis. Conversely, if you get a statistically significant p-value for rejecting the null hypothesis, then you know that k > 1. This strategy can be easily generalized to the multi-variate case by performing PCA and testing each principal component separately.

Since you're working with R, I recommend you take a look at the nortest package.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.