tag:blogger.com,1999:blog-1740670447258719504.post7291605571116925077..comments2024-03-27T03:00:27.024-04:00Comments on Facts & other stubborn things: Ryan Long on the econometrics of the minimum wage: a big picture explanationEvanhttp://www.blogger.com/profile/12259004160963531720noreply@blogger.comBlogger7125tag:blogger.com,1999:blog-1740670447258719504.post-25705611755862027262014-01-19T20:45:14.936-05:002014-01-19T20:45:14.936-05:00Right. But if there's no bias and you add it y...Right. But if there's no bias and you add it you only loose efficiency. If you add it and it flips the sign that suggests something's going on.YouNotSneaky!https://www.blogger.com/profile/06378267534638281151noreply@blogger.comtag:blogger.com,1999:blog-1740670447258719504.post-1645177716981317882014-01-19T15:49:30.437-05:002014-01-19T15:49:30.437-05:00Why shouldn't it flip the sign? If there is bi...Why shouldn't it flip the sign? If there is bias in the original estimate, removing that bias could very well change the sign.Daniel Kuehnhttp://www.factsandotherstubbornthings.blogspot.comnoreply@blogger.comtag:blogger.com,1999:blog-1740670447258719504.post-86151780923291670112014-01-19T15:48:34.523-05:002014-01-19T15:48:34.523-05:00I don't know if I understand what you're t...I don't know if I understand what you're trying to say. Chuck what variable in? Make it complete in what sense? <br /><br />"Policy" is a up to institutions, not the scientist. Daniel Kuehnhttp://www.factsandotherstubbornthings.blogspot.comnoreply@blogger.comtag:blogger.com,1999:blog-1740670447258719504.post-72045028501548350152014-01-19T12:13:34.006-05:002014-01-19T12:13:34.006-05:00So ... chuck in a variable (just to make sure your...So ... chuck in a variable (just to make sure your model is complete), make it "policy", and then record the results. Is that the sum of it? (after all, you are an empiricist rather than a theorist, how else would you test the model?)Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-1740670447258719504.post-86847167129428348722014-01-19T12:11:41.162-05:002014-01-19T12:11:41.162-05:00So ... chuck in a variable (just to make sure your...So ... chuck in a variable (just to make sure your model is complete), make it "policy", and then record the results. Is that the sum of it? (after all, you are an empiricist rather than a theorist, how else would you test the model?)Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-1740670447258719504.post-3990045817871034642014-01-19T06:37:43.920-05:002014-01-19T06:37:43.920-05:00Isn't this the idea that if you add in extra ...Isn't this the idea that if you add in extra stuff into your regression that can make you loose significance but it shouldn't flip the sign of the estimated coefficient (which is what happens in DLR I believe)?YouNotSneaky!https://www.blogger.com/profile/06378267534638281151noreply@blogger.comtag:blogger.com,1999:blog-1740670447258719504.post-37392190184052196802014-01-18T23:07:26.437-05:002014-01-18T23:07:26.437-05:00Thanks for trying to tackle my concerns. You haven...Thanks for trying to tackle my concerns. You haven't really addressed them, but that's my fault for losing interest in making my point, not your fault for not understanding my point.RP Longhttps://www.blogger.com/profile/15028013805248797978noreply@blogger.com