PPPM 413 Study Guide - Final Guide: Dependent And Independent Variables, Statistical Parameter, Stata

• School: University of Oregon
• Department: Planning, Public Policy and Management
• Course: PPPM 413
• Professor: Grant Jacobsen

Document Summary

If regress x -> y (cid:271)ut don"t (cid:272)ontrol for w, then the (cid:272)oeffi(cid:272)ient on x is biased: bias phrasing, absolute values, overestimate: exp. value of coef. > true parameter in abs. value: underestimate: exp. value of coef. < true parameter in abs. value: actual values, positively-biased: exp. value of coef. > true parameter (both +/-: x < (+/-) > w < (+/-) > y, negatively-biased: exp. value of coef. < true parameter (mix of +/-: equation, x < (-/+) > w < (+/-) > y, (cid:1877)=(cid:1853)+(cid:1854)(cid:2869)(cid:1876)(cid:2869)+(cid:1854)(cid:2870)(cid:1876)(cid:2870)+(cid:1854)(cid:2871)(cid:1876)(cid:2871)+(cid:1854)(cid:2872)(cid:1876)(cid:2872)+ , dependent variable = independent variables (most important and other included factors) + variables not in model (all excluded factors, residual) = the impact of being in the corresponding category relative to the omitted category. Indicates the range in which the true population parameter falls: 95% ci = 95% likelihood of the population falls within this range.