1
$\begingroup$

I have the following problem: I have a timeseries with the prices for a few futures, which is non-stationary (according to ADF test). If I apply first difference of logs, ADF shows stationarity. But I also have one column in timeseries with news sentiment scores (from -1 to 1), and it seems like I can't apply first difference of logs to it. This timeseries will be later used in ARIMAX. What should I do? I think all dataset should be in the same I(1) or I(0) condition.

$\endgroup$
2
  • 1
    $\begingroup$ Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking. $\endgroup$
    – Community Bot
    May 13, 2023 at 13:24
  • 1
    $\begingroup$ Note that the first difference of log prices are returns so what you have found is that the calculated return series is stationary. Not that that necessarily matters but I wasn't sure if you were aware of that. If you were, then apologies for noise. $\endgroup$
    – mlofton
    May 14, 2023 at 4:53

2 Answers 2

1
$\begingroup$

I agree with Richard; you don't need to take the log of the (sentiment) negative numbers. Just pass the values of the sentiment factors time series as an exogenous factor in the ARIMAX(.) model.

$\endgroup$
1
  • 1
    $\begingroup$ Your answer could be improved with additional supporting information. Please edit to add further details, such as citations or documentation, so that others can confirm that your answer is correct. You can find more information on how to write good answers in the help center. $\endgroup$
    – Community Bot
    May 23, 2023 at 18:27
2
$\begingroup$

You cannot take logs of negative numbers, but you also do not need to. Use sentiment scores as they are.

$\endgroup$

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.