Theil's Forecast Accuracy Coefficient: A Clarification

By F. Bliemel. 1973

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

A deep learning framework for financial time series using stacked autoencoders and long-short term memory (2017, neutral): “If R is bigger, it means that the predicting value is similar to the actual value, while if MAPE and Theil U are smaller, this also indicates that the predicted value is close to the actual value [23, 76].”
Residential probabilistic load forecasting: A method using Gaussian process designed for electric load data (2018, methodology): “The adoption of the Theil is sometimes accompanied by confusion since there are two versions of the Theil, and each has its own interpretation and parameters [39].”
Construction Sector Contribution to Economic Stability: Malaysian GDP Distribution (2021, methodology): “A value of Theil equal to 1 indicates that the forecasting model is the same as the analyzed model [77].”
Error Measures for Generalizing About Forecasting Methods: Empirical Comparisons (1992, methodology): “Theil's U2 [Theil (1966, ch. 2)] compares the RMSE for a proposed model with the RMSE for the random walk [see Bliemel (1973) for a discussion of Theil's measure].”