I would like to make a comparison on the performance of some regression algorithms according to different performance criteria, including Root Mean squared Error (RMSE), coefficient of determination (R2), and Mean Absolute Percentage Error (MAPE). I found a problem with MAPE.
For example, target values and predicted values correspond to t and y, respectively.
t=[1 , 2 , 3 , 4, 5, 6, 7, 8, 9, 10]
y=[2 , 1, 3 , 4.5 , 5 , 7 , 7, 8 , 9 ,12]
The following performance criteria are obtained:
MAPE: 19.91
RMSE: 0.85
R2: 0.91
While RMSE and R2 are acceptable, the MAPE is around 19.9%, which is too high. My question is that what is the main reason for this high value of MAPE?(compared to acceptable values for RMSE and R2 )
Thanks in advance