1

I want forecast inflation using ARIMA model. My questions are:

  1. Does ARIMA require normally distributed input data? (Because my data—inflation—is not normal.)
  2. If ARIMA require normally distributed input data, how can I transform it to be normal? My data (inflation) contains negative values so I couldn't use the log transform to handle it.
gung - Reinstate Monica
  • 132,789
  • 81
  • 357
  • 650
Leila
  • 11
  • 2

1 Answers1

0

NO. The errors resulting from the ARIMA model must be normally distributed.

John C. Pickett
  • 11
  • 1
  • 3
    Please add the following caveat "for parametric tests of sufficiency( t values for the coefficients) and necessity (randomness of the residuals) to be valid" – IrishStat Aug 01 '15 at 21:29
  • 2
    I would be a bit more careful. It is usually easiest when the errors are normal, but the ARIMA model may still be estimated consistently when they are not. http://math.unice.fr/~frapetti/CorsoP/Chapitre_4_IMEA_1.pdf – Zachary Blumenfeld Aug 01 '15 at 21:30
  • @ZacharyBlumenfeld, nice lecture notes under the link! – Richard Hardy Aug 02 '15 at 08:27