Scalable logistic regression with crossed random effects

HIGHLIGHTS

• who: Swarnadip Ghosh and collaborators from the Department of Statistics, Stanford University have published the Article: Scalable logistic regression with crossed random effects, in the Journal: (JOURNAL)
• what: By properly accounting for crossed random effects the authors show that a naive logistic regression could underestimate sampling variances by several hundred fold. In this model the xij are nonrandom, either because they were designed, or more usually because the analysis is conditional on their observed values. A straightforward solution of equation costs O((R + C + p)3 ) which is infeasible and so the authors develop a . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or E-mail:

  If you don't have an account, you can create one here.