simr

simr copied to clipboard

A small pilot study may not give a precise estimate of data variability. If the data from the larger study turns out to be substantially larger, then realised power may...

pitakakariki

Check if it's still giving different results in some cases. AGHQ=0 initialisation vs last MLE initialisation.

pitakakariki

Power is either 100% or 0%

16

Hello, My colleague and I have been running different variations of a power analysis and run into an issue where the power is either 100% or 0%. Here is our...

ddscott2

cannot use powerSim/doTest when p-values in coefficient results are not labelled "Pr(>|z|)"

1

Hi there, I've fitted a model using geem2, and it outputs the p-values in the coefficients table as simply "p", not "Pr(>|z|)". This results in an error as powerSim/doTest only...

a-ndreaaaaa

Missing Data

3

Either 1. Handle missing data properly. 2. Fail nicely when there is missing data.

pitakakariki

See e.g. https://github.com/pitakakariki/simr/issues/205 Give an error message, suggest use of `extend`. Optional: Automatically extend?

pitakakariki

powerCurve needs to print error/warning counts

3

`powerCurve` and `powerSim` should both probably throw a warning if there are any errors or warnings to report.

pitakakariki

Influence of one NA cell and available options

2

Hello, I would like to report a possible bug, and I would be very grateful for advice to work around it. The issue regards missing data, and I have not...

pablobernabeu

Verbose option for interactive use. Messages when it fits reduced models, print anova tables before extracting p-values.

pitakakariki

This is really just `compare` because there's no random part, but no reason it shouldn't still work.

pitakakariki