where? I saw people who have to use traditional regression for binary

送交者: mangolasi 于 2007-04-18, 16:26:18:

回答: Traditional regression does not make any sense in this case 由 HunHunSheng 于 2007-04-18, 16:22:42:

dependent variable because they use synthetic/aggregated data, and in that case simply b/c E(f(x)) != f(E(x)) for nonlinear f(x), they can no longer logistic/probit regression. Linear regression can perserve the nice property of aggregation.

所有跟贴:

No friggin expected value for binary response - HunHunSheng (32 bytes) 2007-04-18, 16:28:59 (136383)
- you can do a simple linear regression with binary response - xj (43 bytes) 2007-04-18, 18:05:39 (136393)

加跟贴