topic Multiple regression or Two-way ANOVA in Discussions

Multiple regression or Two-way ANOVA

ReyrieNeedshelp — Mon, 04 Mar 2024 11:56:22 GMT

Hello!
I have set of data that includes different cell groups, cultured over 16 days. the cell count is measured every 2 days (sampling days)
I'm looking at multiple regression to check the difference in the cell group*sampling days. I have a total of 9 sampling days and 7 cell groups. my problem is that my parameter estimates do not show the last sampling period (eg. cellgroup1*day 16, cellgroup2*day 16). in fact all cellgroup*day16 does not show up. do you have idea how to solve this issue please and how can I check for significance between cell groups in each sampling date without having to make dummy variable for each cell group and each sampling date.

Can I use two-way ANOVA instead? my residual plot looks like this, so I'm not sure if this violates the assumption of ANOVA as I have a control group that has very high cell count

it would be a lifesaver if someone can answer me. Thank you so much!

Re: Multiple regression or Two-way ANOVA

SamGardner — Tue, 05 Mar 2024 09:26:22 GMT

I think you can get what you want if you use the Effect Leverage personality in Fit Model and look at the LS Means Plot for the interaction term in your model.

Re: Multiple regression or Two-way ANOVA

MRB3855 — Mon, 04 Mar 2024 16:01:35 GMT

Hi @ReyrieNeedshelp You May want to look at expanded estimates so you can view all parameter

estimates.

Details here: https://www.jmp.com/support/help/en/17.2/index.shtml#page/jmp/expanded-estimates.shtml