Question

Regression techniques could involve the issue of 'heteroscedasticity' meaning unequal scatter. This issue is important only when regression is used for demand estimation and not when we use the technique for demand forecasting. Select one:

True

False

Answer 1

So as we know heteroscedasticity is all about change in the spread of its residuals over the range of the measured values from population and when the distribution of the errors in a sample remains constant regardless of its measured value, then, in that case, we called the assumption as homoscedasticity and when it gets violates then it is known as heteroskedasticity so in simple words, we can say even if there is an increase in the observed value, errors should not get increased with it i.e there should not be any direct proportional relationship between an observed value and error value that will be calculated, and as we can see for demand estimation we need to calculate the difference in the mean of the variance between the sample value and the sample mean variance and even for better estimation we can also calculate the difference in the mean of the variance within the observed value, thus definitely heteroscedasticity has an impact on demand estimation whereas in the case of demand forecasting we generally use the historical data thus they don't have any relationship with the observed sample and this is why here there is no impact of heteroscedasticity.

Answer - True