0

if we have two models M1 and M2: MAPE_{M1} = 100% and MAPE_{M2} = 70%$

Does that mean that M2 is better than M1? can we say that M2 is (100/70 -1)*100 = 43% better?

lalaland
  • 237

1 Answers1

2

Certainly you can say that. You could also say that M2 is $1-\frac{\text{MAPE}_{M2}}{\text{MAPE}_{M1}}=30\%$ better. It depends on what the denominator of your percentage calculation of "better" is.

However, whether all this has a useful meaning is very doubtful. As is the entire notion of MAPE. It looks easy to interpret and compare (which it isn't, this impression is highly misleading), but I have never seen an actual business problem that would best be addressed by a forecast that minimized MAPE, rather than one that minimized MSE or a pinball loss.

You may want to take a look at What are the shortcomings of the Mean Absolute Percentage Error (MAPE)?

Stephan Kolassa
  • 123,354