1. 𝐀 (g-1) , 𝐁 (g-1) , and 𝚺 ? ? (g-1) , and given the choice results 𝐲, update the augmented utility function in differences by sampling ? 1 𝐔 (g) ~𝒯𝒩 R|𝐲 (𝐗𝐀 + 𝐳 U 𝐁,𝚺 ? ? ), where R is the truncation region defined by max? 1 𝐔 𝒾 ? 0 if 𝐲 = 1, or by ? 1 𝐔 𝒾 > max{0,? 1 𝐔 -𝒾 } if 𝐲 > 1;Conditional on 𝐳 (g)
2. Bayesian estimation of mixed logit models: Selecting an appropriate prior for the covariance matrix;D Akinc;Journal of choice modelling,2018
3. Bayesian estimation of mixed multinomial logit models: Advances and simulation-based evaluations;P Bansal;Transportation Research Part B: Methodological,2020