nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

594194_en-US

liliconcom

Question about marginal and conditional residuals

Hey all!

I got this question about marginal and conditional residuals from a customer onsite. I wanted to see if anyone had thoughts on this. It's got me stumped.

My question related to the use of marginal and conditional residuals – will JMP’s fit model support the use of marginal and conditional residuals in the analysis diagnostics. Currently, it only plots marginal residuals, fixed effects only, as opposed to the conditional residuals – which includes the fixed effects and random effects. The question is can the output for the residuals support changing them or identifying them so that the average user can have access to it and be able to interpret the diagnostics correctly.

Thoughts?

Best,

Re: Question about marginal and conditional residuals

That's my guess, yes - the issue appears to be around the use of random block (and REML in Standard Least Squares - Base JMP). It looks like you can output the Conditional Residuals as a column, but only Fit Mixed (Pro) has the conditional residual plot.

Re: Question about marginal and conditional residuals

I assume when they say "Fit Model" they're referring to standard least squares, because Mixed Model within Fit Model does provide conditional diagnostics?

Comments

There are no comments yet