[Feature Request]: Clustered standard errors for OLS regression

[bobreednz]

Description

It would be great if one could have clustered standard errors for linear regression

Purpose

Meta-analysts sometimes include OLS regressions along with FE/RE, etc. CR1/CR2 standard errors are available for those estimators, so it would be nice to have them for OLS as well to faciliate comparison

Use-case

Don't know what this means

Is your feature request related to a problem?

Not a problem. This is a suggestion for an added feature.

Is your feature request related to a JASP module?

Regression

Describe the solution you would like

An option for CR1 and CR2 clustered standard errors

Describe alternatives that you have considered

No response

Additional context

No response

[JohnnyDoorn]

Hi @bobreednz,

Just to clarify - did you take a look at the meta-analysis module for conducting such analyses, or do you mean that you would like this feature also for regular (non-meta-analysis) regression analyses?

Cheers
Johnny

[TarandeepKang]

Hi Johnny, I think this is a request for a new feature in regression analysis? I only comment here because if this is the case, it then overlaps with my older request: #2213 Although there motivated by improved interpretability of The coefficients themselves. Moreover, it does look like a very substantial overlap in the underlying theory. Both rely heavily on the delta and sandwich packages as far as I can tell, too?

Best,

Tarandeep