12.2 Performance Measurements

12.2.1 Error terms

  • Mean error (ME):

ME=1n(YtˆYt)$

  • Mean Absolute Deviation (error):

MAD(i.e. MAE) = 1n|YtˆYt|

  • Mean Percentage Error (MPE):

MPE = 1n (YtˆYt)Yt

  • Mean Absolute Percentage Error (MAPE):

MAPE = 1n |(YtˆYt)||Yt|

  • Mean-Squared Error (MSE):

MSE=1n(YtˆYt)2

  • Root Mean-Squared Error:

RMSE=MSE

12.2.2 Multicollinearity

VIF

VIFj=11R2j

Where j=1,...,k

Thus, we see that Rsquare is obtained from regression each IDV against the remaining variables. We can then have the following outputs:

  • VIF = 1, no milticollinearity
  • VIF > 10, indicates multicollinearity

If one gets an indication of multicollinearity, then one should drop one of the correlated variables.