标题: T-stat for multiple regression [打印本页] 作者: badonkus 时间: 2011-7-13 16:20 标题: T-stat for multiple regression
A little late in the game to be asking this, but I assumed I had the basics of quant (at least compared to some sections of PM/Econ)...
Anyway, Stalls indicates the formula to solve for t-stat of regression coefficient is:
Tcalc = b1-B1/sb1
CFAi indicates the formula is simply b1/sb1
If you need a specific example I am referring to #40 of the CFAi PM mocks.作者: chaojimali 时间: 2011-7-13 16:20
It the man ( B1 in the above ) is zero , then CFAI is same as Stalla . Read the question carefully作者: ppls 时间: 2011-7-13 16:20
Thanks for the quick response. I'll go back and read the question, but I thought for sure they indicated the null was = 1, not zero in this case.作者: Swanand 时间: 2011-7-13 16:20
The general formula is Tcalc = b1-B1/sb1
The T stat that you see in the ANOVA tables is usually based on B1 =0. Hence, the formula is then: Tcalc = b1/sb1
