tag:blogger.com,1999:blog-5624436327404149621.post4325667628886726428..comments2024-03-28T20:07:31.640-07:00Comments on Quantitative Ease by Carola Binder: Thoughts on the Fed's New Labor Market Conditions IndexCarolahttp://www.blogger.com/profile/12783977056485775882noreply@blogger.comBlogger8125tag:blogger.com,1999:blog-5624436327404149621.post-58160986709365808232021-11-13T04:56:16.097-08:002021-11-13T04:56:16.097-08:00yeezy 350<a href="http://www.yeezy-350.us.com" rel="nofollow"><strong>yeezy 350</strong></a><br />jasonbobhttps://www.blogger.com/profile/05953248680938378417noreply@blogger.comtag:blogger.com,1999:blog-5624436327404149621.post-85573897178343224842016-10-19T03:59:56.099-07:002016-10-19T03:59:56.099-07:00Ah yes, I got in contact with the Fed economists a...Ah yes, I got in contact with the Fed economists and I've managed to replicate it. You need to take the rolling annual change in the LMCI and regress it against the annual change in the unemployment rate. I get an r-squared of about 0.9, give or take.<br /><br />I also ran a test to see if either series leads the other. I just did it informally with a simple lag and didn't bother with a Granger test. But the correlation drops, so it is clear that neither series predicts the other.<br /><br />Given all that I think that your post was correct: the LMCI is a pretty pointless indicator. We're probably better off just looking at annual changes in the unemployment rate and then looking at other indices in order to form a deeper understanding of various aspects of the labour market.<br /><br />One last comment: the LMCI starts to deteriorate at the point at which the fall in the unemployment rate starts to decelerate. That is, when the economy begins to reach the top of the employment cycle. I find that interesting. Not sure if you will.TheIllusionisthttps://www.blogger.com/profile/17642837989235595346noreply@blogger.comtag:blogger.com,1999:blog-5624436327404149621.post-23469142781617135422016-10-19T03:59:45.640-07:002016-10-19T03:59:45.640-07:00This comment has been removed by the author.TheIllusionisthttps://www.blogger.com/profile/17642837989235595346noreply@blogger.comtag:blogger.com,1999:blog-5624436327404149621.post-50661369388279724912016-10-19T03:59:26.960-07:002016-10-19T03:59:26.960-07:00This comment has been removed by the author.TheIllusionisthttps://www.blogger.com/profile/17642837989235595346noreply@blogger.comtag:blogger.com,1999:blog-5624436327404149621.post-21739968544843390582016-10-17T18:06:42.883-07:002016-10-17T18:06:42.883-07:00I took it from the first row, last column of the t...I took it from the first row, last column of the table at this link: http://www.federalreserve.gov/econresdata/notes/feds-notes/2014/assessing-the-change-in-labor-market-conditions-20140522.htmlCarolahttps://www.blogger.com/profile/12783977056485775882noreply@blogger.comtag:blogger.com,1999:blog-5624436327404149621.post-38314508888971623162016-10-17T09:13:13.692-07:002016-10-17T09:13:13.692-07:00Hi,
I'm having a very hard time replicating y...Hi,<br /><br />I'm having a very hard time replicating your claim of a 0.96 correlation coefficient.<br /><br />If I regress raw unemployment numbers on raw LMCI I get a positive slope and an r-squared of 0.0023.<br /><br />So I played around and the best fit I could get is if I take MoM change in the unemployment and regress it against the raw LMCI. If I do this I get a negative slope and an r-squared of 0.2925.<br /><br />Not a terrible correlation but far from your reported number.<br /><br />Could you please either lay out your methodology and data source. Or even just post some screens of a scatterplot with a best fit line.<br /><br />Thanks,<br /><br />PTheIllusionisthttps://www.blogger.com/profile/17642837989235595346noreply@blogger.comtag:blogger.com,1999:blog-5624436327404149621.post-92129710135265270112014-09-02T03:17:05.145-07:002014-09-02T03:17:05.145-07:00great post! factor models are attractive but their...great post! factor models are attractive but their usefulness depends partly on what they're used for. one of the differences in factor and component extraction is the source of common variance (communalities/uniqueness). it would be nice to see the extraction details as well as the loadings/rotations used to better interpret the signals that each variable send out through the index.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5624436327404149621.post-7491437112550515432014-07-18T06:43:33.287-07:002014-07-18T06:43:33.287-07:00I can see the usefulness of the index (sans wages ...I can see the usefulness of the index (sans wages for the reasons you cite) as a sort of defense from those who are always saying, "yes but the REAL unemployment rate is XX." Thaomashttps://www.blogger.com/profile/14747215297590200584noreply@blogger.com