search

rondolab / MR-PRESSO

Performs the Mendelian Randomization Pleiotropy RESidual Sum and Outlier (MR-PRESSO) method.
81 stars 30 forks source link

Output SD or SE? #15

Open btaschler opened 3 years ago

btaschler commented 3 years ago

Hi, The fitted object contains a field "Sd". Could you please clarify whether these are standard deviations or rather standard errors of the effect estimate? Thanks!

kaanokay commented 2 years ago

Hi @rondolab ,

I have the same question. Should we should provide standard error of beta or standard deviation of beta? Could you clarify it?

Thanks. Bw.

kaanokay commented 2 years ago

Hi, The fitted object contains a field "Sd". Could you please clarify whether these are standard deviations or rather standard errors of the effect estimate? Thanks!

Hi @btaschler do you have answer for your question? I have the same question with you. If you know this now, could you share with me?

Thanks!

btaschler commented 2 years ago

Hi @kaanokay, No, I didn't get a response so far. My best guess is that it should be the standard error (in the sense that "sd" refers to the standard deviation of the mean which would be the standard error), but I'm not 100% certain. Cheers

kaanokay commented 2 years ago

Hi @btaschler,

Thank you for your quick reply. I'm not sure also. Do you know another approach to test pleiotropy?

Bw.

marieverbanck commented 2 years ago

Hi @btaschler and @kaanokay, sorry for our late reply, sd stands for the estimated standard error of the causal estimate. Best, Marie

kaanokay commented 2 years ago

Hi @marieverbanck,

Thank you for reply.

Bw.

marynias commented 1 year ago

I would change the column name to "se" as "sd" is misleading as it is a standard abbreviation for "standard deviation"....