“…N is the sample size of profiles and j = 1,…, M . As shown in Shang et al , the collected observations are assumed from the following profile model: where α j is the intercept parameter and β j =( β 1 j ,…, β q j ) T is a q ‐dimensional parameter vector. Here, y j i is assumed to be drawn from a binomial (or Bernoulli) distribution with the parameter p j i , denoted y j i ∼Binomial( n j i , p j i ).…”