Hi all:
I am using PSM to test whether the treatment effect is siginificant. In the psm command, I included four varibles in outcome option, like psmatch2 treat x1,x2,x3,outcome (y1 y2 y3 y4). The psmatch2 command can give me four varibales' treatment effect respectively. However, when I use bootstrap to calculate a more accurate SE, the values of ATT,ATE and ATU apear once only, the command is like: bootstrap r(att) r(ate) r(atu) : psmatch2 treat x1,x2,x3,outcome (y1 y2 y3 y4). My question is How should I interpret the result of bootstrap when including multiple variables at the same time? And if I want to calculate SE for each variable, should I only include one variable each time when using bootstrap?
Thanks a lot for helping me understand this question.
I am using PSM to test whether the treatment effect is siginificant. In the psm command, I included four varibles in outcome option, like psmatch2 treat x1,x2,x3,outcome (y1 y2 y3 y4). The psmatch2 command can give me four varibales' treatment effect respectively. However, when I use bootstrap to calculate a more accurate SE, the values of ATT,ATE and ATU apear once only, the command is like: bootstrap r(att) r(ate) r(atu) : psmatch2 treat x1,x2,x3,outcome (y1 y2 y3 y4). My question is How should I interpret the result of bootstrap when including multiple variables at the same time? And if I want to calculate SE for each variable, should I only include one variable each time when using bootstrap?
Thanks a lot for helping me understand this question.
