mguindani.bsky.social
Statistician. Bayesian. Professor, Department of Biostatistics, University of California, Los Angeles. Views are my own. He/Him.
128 posts
1,196 followers
927 following
Prolific Poster
comment in response to
post
😆
comment in response to
post
Cancel and dictate culture
comment in response to
post
We have previously shown in context of autoregressions with exogenous regressors that the use of joint log-score in cross-validation reduces the variance of the CV estimate leading to better efficiency in model comparison doi.org/10.1214/23-B..., and in the new paper we extend this to spatial models
comment in response to
post
Concentration camps in another country. Not really much else to say.
comment in response to
post
How is it an extension of the US prison system if people can be brought in, there is no due process, and if a mistake is made, it looks like they are lost forever? It looks like the US is simply paying money for foreign concentration camps.
comment in response to
post
Well, the a priori probability of beating the market is not high. If the probability is about 10% for professional fund managers, for regular investors, that would likely be around 1%. Then, the (post-pred) probability of an outcome >=35 is just 0.26. Just saying 🙃
comment in response to
post
A failure to reject the null is not an acceptance of the null.
comment in response to
post
?? Like that percentage of yours? 😀
comment in response to
post
A losing company paying off the debts of another losing company.