# ARIMA with increasing variance

my time series data looks like that:

I would like to use ARIMA model to forecast next steps in time series. Unfortunately because of increasing variance data is non-stationary. Is there any way to transform my data to get rid of this variance? In addition before training I normalised data. Is normalisation required by ARIMA model?

• Try a variance stabilizing transform, like the log. – Demetri Pananos Jan 31 at 5:51
• What demetri said is fine. You could also try to model the process directly by saying possibly $y_t = \mu + \epsilon_t$ where the variance of $\epsilon$ is say (arch or garch ) or stochastic ( so SV model ). Just another possibility. – mlofton Jan 31 at 6:20

As Demetrios says, you could try transforming your data to stabilize the variance. The classical approach is to use a Box-Cox transformation. One special case of this is the log transform. Some software will estimate a Box-Cox transformation parameter $$\lambda$$ for you.