I show how conditional Gaussians, whose means are conditioned by a random variable, can be estimated and their likelihoods computed. This is based upon how regular Gaussians have their own parameters and likelihood computed. After explaining how to estimate the parameters of Gaussians and conditional Gaussians, I explain how to calculate their likelihoods even if there are missing elements in the data or, in the case of the conditional Gaussian, even if the conditioning variable is missing.