Comments on: A Quote on Model http://hea-www.harvard.edu/AstroStat/slog/2008/a-quote-on-model/ Weaving together Astronomy+Statistics+Computer Science+Engineering+Intrumentation, far beyond the growing borders Fri, 01 Jun 2012 18:47:52 +0000 hourly 1 http://wordpress.org/?v=3.4 By: Alex http://hea-www.harvard.edu/AstroStat/slog/2008/a-quote-on-model/comment-page-1/#comment-785 Alex Thu, 09 Oct 2008 02:12:13 +0000 http://hea-www.harvard.edu/AstroStat/slog/?p=984#comment-785 It appears to me that many inference problems in astronomy are essentially variants of <a href="http://en.wikipedia.org/wiki/Generalized_linear_model" rel="nofollow">generalized linear models</a> (of the Poisson distribution, log link variety, in particular). Granted, most of the problems we work on in the astrostatistics group are more complex due to a large number of parameters or other complications. However, for everyday astronomical analysis, GLMs seem like a natural improvement on linear regression and other methods based on Gaussian error models. It appears to me that many inference problems in astronomy are essentially variants of generalized linear models (of the Poisson distribution, log link variety, in particular). Granted, most of the problems we work on in the astrostatistics group are more complex due to a large number of parameters or other complications. However, for everyday astronomical analysis, GLMs seem like a natural improvement on linear regression and other methods based on Gaussian error models.

]]>