Research Article

An Advanced Hybrid Forecasting System for Wind Speed Point Forecasting and Interval Forecasting

Table 4

The indexes for evaluating the prediction ability.

MetricDefinitionEquation

MAPEThe average of N absolute percentage error
MAEThe mean absolute error of N forecasting results
RMSEThe square root of the average of the error squares
MSEMean squared error
DCDirectional change
FICPForecast interval coverage probability
FINAWForecast interval normalized average width
AWDAccumulated width deviation of testing dataset

In the formulas, Li and Ui represent the lower and the upper limit of the production interval, respectively. And, ci is the number of the truth value contained in constructed interval. N is the number of the testing set [16]. Data are the actual data. Besides, AWDi is the width deviation of the construction interval of each sample.