We are reporting the difference between the point estimate and counterfactual at time t+4 [4 quarters after intervention].
Fit a glm Poisson withibtervention and time since intervention dummy variables.
How do I derive a specific p-value for the outcome at t+4?
David Eugene Booth
Take a look at these two papers they may be of some help. Best wishes David Booth
0 votes
0 thanks
John Rahilly
Thanks David,
I've looked at these papers and others, but still can't arrive at an understanding of how anyone derives a p-value for a specific point estimate difference at a point later than the model output. Maybe I'm missing something?
0 votes
0 thanks
David Eugene Booth
John, it's called extrapolation and is not ever recommended for regression models, Because estimates outside your data set are extremely risky . Best wishes David Booth
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size
More John Rahilly's questions See All
Similar questions and discussions