Title :
A new measure of forecast accuracy
Author :
Chen, Haijun ; Wu, Lifeng
Author_Institution :
Dept. of Math., Handan Coll., Handan, China
Abstract :
In this paper, we present a new measure of forecast accuracy: a modified mean absolute percentage error (MAPE). As a means of establishing the goodness of fit, we set different objective functions, by implementation of a global optimization algorithm, the results of setting the objective function as modified MAPE are compared with the results of setting the objective function as MAPE, comparisons demonstrate that modified MAPE comes with lower prediction error.
Keywords :
error analysis; forecasting theory; mean square error methods; optimisation; forecast accuracy; global optimization algorithm; mean absolute percentage error; Accuracy; Data models; Forecasting; Mathematical model; Optimization; Predictive models; Forecasting; MAPE; Modified MAPE;
Conference_Titel :
Information and Financial Engineering (ICIFE), 2010 2nd IEEE International Conference on
Conference_Location :
Chongqing
Print_ISBN :
978-1-4244-6927-7
DOI :
10.1109/ICIFE.2010.5609455
