In statistics, a sequence or a vector of random variables is heteroscedastic if the random variables in the sequence or vector may have different variances. The converse is called homoscedasticity.