tag:blogger.com,1999:blog-2474498300859593807.post776347916080924440..comments2024-03-08T02:53:17.969-05:00Comments on Econometric Sense: Data science = failure of imaginationMatt Bogardhttp://www.blogger.com/profile/10510725993509264716noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-2474498300859593807.post-70985859052711747242013-01-09T08:13:39.856-05:002013-01-09T08:13:39.856-05:00Nice article!
Nice article!<br /><br />Philhttps://www.blogger.com/profile/13523323134340990902noreply@blogger.comtag:blogger.com,1999:blog-2474498300859593807.post-9180960086049825222013-01-09T01:10:25.959-05:002013-01-09T01:10:25.959-05:00Sure, understanding the mechanics and theory behin...Sure, understanding the mechanics and theory behind data mining / machine learning algorithms is difficult. No one should say differently. The question you should be asking is whether your average "data scientist" actually knows or even cares about the theory.<br /><br />My, albeit limited, experience with data scientists has led me to believe that the majority blindly apply whatever algorithm is in vogue... (There are a lot of statistical quacks out there!) Many of them don't even think twice about what the data generating process might be... Of course, that's sometimes beneficial. This is esp. true in a commercial environment, where the main concern is prediction / classification rather than understanding of the DGP. <br /><br />Contrast that to the case of econometric research / analysis, where your main concern is usually about understanding the dynamics... <br /><br />Different tools for different tasks.Anonymoushttps://www.blogger.com/profile/15529961438901511416noreply@blogger.com