search

elmacartney / EE_stress_MA

MIT License
0 stars 0 forks source link

moderator weights matching marginal R2 #15

Closed elmacartney closed 3 years ago

elmacartney commented 3 years ago

The moderators with the highest weights don't seem to also have the highest marginal R2.

Also, with the modified habituation data, all stress models have an AIC > 2 compared to the best fitting model so the importance function gives this error: "Error in sw.model.selection(res_Sfm2) : argument consists of only one model" unless I increase the AIC threshold

itchyshin commented 3 years ago

This is OK - that we only have one model within delta AIC and you can go on @elmacartney

elmacartney commented 3 years ago

@itchyshin could you please take a look at table 4 of the manuscript when you get a chance and see what you think about the weights vs the R2. I know that the correlations can affect them but there doesn't seem to be any clear consistency between them.

If you want to look at the correlation figure, it is on the html "Dataset_summary" on github

itchyshin commented 3 years ago

@elmacartney - thanks for letting me know - I wanted to check R2 and weights stuff anyways so I shall do this as soon as I can. You can certainly start writing results based on uni-moderator models tho.