0

So I already ran some tests to make my data stationary. Differencing and box-cox transformation in particular. According to the augmented-dickey fuller test, after performing the above mentioned transformations, the data does not have a unit root but it is still not normally distributed. Can someone enlighten me about this?

mpiktas
  • 33,140
  • 5
  • 82
  • 138
thinking
  • 1
  • 1
  • 1
    http://stackoverflow.com/questions/14062635/why-my-boxcox-transformation-does-not-result-a-normal-data – Karel Macek Apr 09 '15 at 08:32
  • Let the two sequences $(\ldots,0,0,0,\ldots)$ and $(\ldots,1,1,1,\ldots)$ have equal probabilities of $1/2$. Each component therefore has a Bernoulli(1/2) distribution--it's not Normal. Is your definition of "stationary" flexible enough to recognize this as a stationary process? If not, exactly what is your definition? – whuber Apr 09 '15 at 16:00
  • Related question http://stats.stackexchange.com/q/38852/6633 – Dilip Sarwate Apr 09 '15 at 16:08
  • 1
    To answer briefly, stationary data need not be normal. Normality is not a requirement for stationarity (regardless of which definition -- strict stationarity or weak stationary -- you use). – Richard Hardy Apr 09 '15 at 18:36

0 Answers0