2
$\begingroup$

I have panel data with COVID-19 variables starting in 2017 to look at how the pandemic affected stock markets. My N = 487 with 494500 daily observations. At the moment I left the time period before 2020 blank in the panel data resulting in my models only taking into account 151000 observations. Am I allowed to fill in the data period from 2017 to 2020 with 0's if I use the daily amount of new vaccinations and deaths to increase the amount of rows incorporated in the model?

Kind regards

$\endgroup$
2
  • $\begingroup$ Just to be clear, what is your outcome of interest? $\endgroup$ Dec 29, 2022 at 2:37
  • $\begingroup$ how the pandemic has affected the liquidity in stock marets @ThomasBilach $\endgroup$
    – SDC
    Dec 30, 2022 at 0:32

1 Answer 1

1
$\begingroup$

Since you are interested in the stock market a more elegant way to incorporate the pandemic covariates would be to include a set of dummy variables (be aware of the dummy variable trap) and, depending on other independent variables you are considering, interaction terms of the pandemic dummy variables and other covariates. If you are dealing with daily data you may also consider to employ Chow test to determine whether there is a structural change between the two different periods.

I suggest you to have a look at Chapter 13 of Introductory Econometrics: A Modern Approach by Jeffrey Wooldridge.

$\endgroup$
1
  • $\begingroup$ Thank you for your suggestion. I'll look into it $\endgroup$
    – SDC
    Dec 29, 2022 at 1:13

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.