0
$\begingroup$

I have these data, related to value of pH over time: data I want to generate a stochastic model (ARMA, ARMAX, and so on), and I want to decide the order of this model, so I did the plots of ACF and PACF, because I know that when the value of PACF goes essentially to 0, that is the order of the model. But I cannot interpret these plots, because I think some pre-processing steps are required. ACF and PACF

Improve this question
$\endgroup$
3
  • $\begingroup$ The data does not look stationary. Did you run a unit root test on the data? Before you do anything else you have to make it stationary. I would run auto.arima in the forecast package and see what it tells you as a starting point. I am confused what you mean by stochastic, the data is either that or its deterministic. You have to know that, there is no statistical way I know to determine this. $\endgroup$
    – user54285
    Commented May 14, 2021 at 22:25
  • $\begingroup$ Ok, right, I will make my data stationary! $\endgroup$
    – CasellaJr
    Commented May 15, 2021 at 7:58
  • $\begingroup$ auto.arima in the forecast package is of MATLAB or R? $\endgroup$
    – CasellaJr
    Commented May 15, 2021 at 8:05

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.