0
$\begingroup$

I have a balanced panel data set containing variable y = daily numbers of COVID-19 reported cases at the municipal level over a 3-month period. I want to know if another daily-measured variable (x) is related to the local COVID situation.

I also have several time-invariant variables (ie, sd = sociodemographic variables) and some variables that vary in time but are constant across areas (ie, cn = the number of reported cases in the whole country). In a classic situation I would perform an OLS linear regression such as

lm(formula = y ~ x + sd + cn, data = df)

However this is a panel dataset. I suppose I could perform a random effects model to assess the influence of local reported cases on my dependent variable, however I also would like to assess the role of the other variables. What would be an alternative to panel data regression?

$\endgroup$
-1
$\begingroup$

This sounds like the use case for a correlated random-effects model:

https://journals.sagepub.com/doi/pdf/10.1177/1536867X1301300105

$\endgroup$
1
  • 4
    $\begingroup$ Welcome to CV! Your answer is too short as it currently stands. Please provide the gist of the link, and also the full reference in case the link dies in the future. Thank you $\endgroup$
    – Antoine
    Oct 23 '20 at 13:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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