2013 · 426 citations · 1 references
BusinessMeteorologyForecastingNumerical Weather PredictionTime Series AnalysisForecast Error AnalysisForecast EvaluationForecast UncertaintyForecasting MethodsMeasurement ErrorTime Series ForecastingAccuracyEvaluationForecast Accuracy AnalysisForecast Performance EvaluationForecasting AccuracyForecasting Accuracy EvaluationMean Absolute Forecast ErrorTime Series Forecast Error
Each forecast error measure has disadvantages that can lead to inaccurate evaluation, making it impossible to choose a single measure and necessitating recommendations for selecting appropriate measures. The article reviews common forecast error measurements and proposes an Integral Normalized Mean Square Error to reduce outlier impact. Formulas for all error measures are presented, their drawbacks discussed, and the Integral Normalized Mean Square Error is introduced. Error measures are grouped into seven categories: absolute errors, percentage errors, symmetric errors, relative errors, scaled errors, relative measures, and other error measures.
This article reviews the common used forecast error measurements. All error measurements have been joined in the seven groups: absolute forecasting errors, measures based on percentage errors, symmetric errors, measures based on relative errors, scaled errors, relative measures and other error measures. The formulas are presented and drawbacks are discussed for every accuracy measurements. To reduce the impact of outliers, an Integral Normalized Mean Square Error have been proposed. Due to the fact that each error measure has the disadvantages that can lead to inaccurate evaluation of the forecasting results, it is impossible to choose only one measure, the recommendations for selecting the appropriate error measurements are given.
1