Viewing posts tagged From old blog
Posted by:
kerem
in
Journal Paper
8 years, 5 months ago
Continuing with the new regression model. Now it's the discrete dataset.
Posted by:
kerem
in
Uncategorized
8 years, 5 months ago
It works for the original dataset pretty well. Here's what comes out:
Posted by:
kerem
in
Uncategorized
8 years, 5 months ago
I have just discovered a better model for participants scores in the experiment. This used to be our prime model:
```prettyprint
model <- lmer(score ~ nstates_amp_and_freq_n + (1 | id) + (1 |condition:phase) , data=all, REML=F)
```
I have discovered the following one outperforms this for both the original and the discrete datasets, and is much easier to interpret:
```prettyprint
model <- lmer(score ~ nstates_amp_and_freq_n:phase_order:phase + (1|id), data=all, REML=F)
```
Of course, we are using MCMC samples, so the model declaration becomes:
I have just noticed that the first journal paper we are planning to send in with Hannah contains a huge error: I keep mentioning predictors, but I should be talking about random effects!
There is a lot of stuff online about eliciting p-values for lme4 linear mixed-effects regression model coefficients (just google it). I opted for MCMCglmm because I like MCMC , and because lme4 has deprecated its own MCMC implementation.