search

firasm / interCLASS

InterCLASS analysis code
0 stars 0 forks source link

Calculate uncertainties for shift maps (differences between pre and post) #5

Open firasm opened 9 years ago

firasm commented 9 years ago

There is a technical reason I haven't been showing uncertainties in difference data.

screen shot 2015-07-06 at 12 59 58 am

The reason is that I'm not actually sure how to calculate/propagate them correctly given that the populations are the same. I have the SEM of the pre and the SEM of the +1 week.

Do I just pretend I'm subtracting independent numbers and combine them as normal? It'll take a bit of time to do that, but I can if you think this is the correct way to show them. I was just going to point to the uncertainties from result 2 to show that the uncertainties are really small.

I'll let @jamescharbonneau comment on this one

jamescharbonneau commented 9 years ago

Yeah, I think you treat them like they're independent numbers.

firasm wrote:

There is a technical reason I haven't been showing uncertainties in difference data. screen shot 2015-07-06 at 12 59 58 am https://cloud.githubusercontent.com/assets/2507459/8517974/aefd71e4-237a-11e5-8e9f-e79530f97726.png

The reason is that I'm not actually sure how to calculate/propagate them correctly given that the populations are the same. I have the SEM of the pre and the SEM of the +1 week.

Do I just pretend I'm subtracting independent numbers and combine them as normal? It'll take a bit of time to do that, but I can if you think this is the correct way to show them. I was just going to point to the uncertainties from result 2 to show that the uncertainties are really small.

I'll let @jamescharbonneau https://github.com/jamescharbonneau comment on this one

— Reply to this email directly or view it on GitHub https://github.com/firasm/interCLASS/issues/5.