Question

Which one of the following statements about Theil's U Stat measure of error is correct? Note: Theil's U Stat error measure is discussed in my measurement error audio file.

U Stat - (Standard error of the forecast model / Standard error of a forecast developed using a Naive Approach) 

Answer 1

The answer is the last option i.e.

If U = 1 the naïve method is as good as the forecasting being used.

If U<1 the forecasting technique being used is better than the naïve method. The smaller U, the better the technique.

If U>1 the naïve method produces better results than the technique being used.

Explanation: Theil’s U statistic is a relative accuracy measure that compares the forecasted results with the results of forecasting with minimal historical data. It also squares the deviations to give more weight to large errors and to exaggerate errors, which can help eliminate methods with large errors. The above interpretations for U = 1, U<1 and U>1 are standard interpretations.