4
$\begingroup$

I am not used to time series econometrics. If I understand well, before conducting a linear regression (the most basic possible), we need to ensure the stationarity of our data - particularly with (Augmented) Dickey-Fuller tests. Then, we need to fit the optimal ARMA(p,q) model to our stationary data.

Once we've fitted ARMA model to the data, we can do forecasts, however I don't really understand how we can use them in linear regressions. Do we need to take the residuals of the ARMA process (our series without AR or MA influence) as our "new" variables for our regression ? Do we loose much information by doing so ?

I understand that we fit the ARMA process on regression residuals to check their stationarity and ensure that we did not do a spurious regression. But I am not sure on how to use the ARMA before the regression !

$\endgroup$
2
  • $\begingroup$ Why not jointly model the ARMA process in a suitable model for dependent data? $\endgroup$
    – AdamO
    Nov 3, 2023 at 16:04
  • $\begingroup$ Doesn't it comes back to the same thing ? Extracting residuals and using them instead of the original data in the regression // using the original data in the regression and adding ARMA process as predictors ? $\endgroup$
    – krauuuus
    Nov 3, 2023 at 16:08

1 Answer 1

4
$\begingroup$

ARMA is a pure time series model, i.e. it doesn't have exogenous variables $X_t$ unlike a typical regression $y_t=X_t\beta+\varepsilon_t$. In a generic formulation of ARMA such as $\phi(B)y_t=\theta(B)\varepsilon_t$ the exogenous variables $X_t$ are nowhere to be seen.

There are a few ways to marry these two models. One is called ARIMAX: $$\phi(B)y_t=X_t\beta+\theta(B)\varepsilon_t$$ and the other is regARIMA: $$y_t=X_t\beta+u_t\\\phi(B)u_t=\theta(B)\varepsilon_t$$

The latter formulation can be easier to understand for you as it states that we have a linear regression where the errors are ARIMA, which is similar to how you described your understanding. Both approaches are estimated in MLE framework where exogenous coefficients $\beta$ and ARMA coefficients are estimated simultaneously, not in the two step approach that you suggest.

If you didn't have moving average part in ARMA, then the ARIMAX can be estimated in a linear regression with OLS: $$\phi(B)y_t=X_t\beta+\varepsilon_t$$ $$y_t=\phi_1 y_{t-1}+\phi_2 y_{t-2}+\dots+X_t\beta+\varepsilon_t$$

$\endgroup$
3
  • $\begingroup$ So if I understand well, to conduct a non-spurrious regression, I must first fit the optimal ARMA model to my time series, then two options : (i) add external regressors with regARIMA or (ii) use the regression of the ARIMA model as my new variables. Is it correct ? $\endgroup$
    – krauuuus
    Nov 5, 2023 at 14:36
  • $\begingroup$ @krauuuus, no. you fit both exogenous and endogenous variables in one shot with a suitable statistical package. for instance, Python stats models package has regARIMA and a limited ARIMAX formulations. As I wrote in some cases you can fit everything in one shot with OLS. there's no need to do two steps like you describe, and it won't give you deserted result anyways $\endgroup$
    – Aksakal
    Nov 5, 2023 at 18:49
  • $\begingroup$ Ok, thank you for your answers ! $\endgroup$
    – krauuuus
    Nov 5, 2023 at 21:43

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.