[讨论]Multilevel MixedTobit/Censored Model?

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

I am planning to estimate a censored model with fixed and random effects at a number of levels and I am trying to find an appropriate software for this.The model is to explain vehicle kilometres travelled by individuals so it is left censored at zero.The model is to contain random effects for individuals, households and neighbourhoods and has fixed effects for the same levels (we are estimating the effects of neighbourhood variables).

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：Multilevel Censored censor model multi software planning contain levels number

http://support.sas.com/documentation/onlinedoc/stat/930/mcmc.pdf

I would not regard that as censoring because the variable "kilometres travelled" is naturally bounded at zero. By regarding this as some form of censoring mechanism one would assume that distance travelled could in principle be negative, but negative values were just not directly observed, for whatever reason. Maybe a GLMM with Gaussian or Poisson errors and a log link would be a more reasonable choice for this kind of data.