Is the condition where the residual variance is nonconstant. This tends to happen in cross-sectional analyses. It also occurs in various consumer, income, risk, and size of firm studies. This compares to Homoskedasticity.
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 complementary concept is called homoscedasticity. (Note: There seems to be no standard agreed-upon spelling for these words; they are sometimes spelled homo- or heteroskedastic or (incorrectly) -schedastic, depending on location and personal taste.)