Talk:Econometrics
From Wikipedia, the free encyclopedia
 |
This article is within the scope of the Economics WikiProject, an effort to create, expand, organize, and improve economics-related articles.. |
| Start | rated as Start-Class on the assessment scale |
| Top | rated as Top-importance on the importance scale |
This is a talk page. Please respect the talk page guidelines, and remember to sign your posts using four tildes (~~~~).
Contents |
[edit] Criticism?
There are plenty of critics of econometrics and its failure to model many aspects of economies. Why is there no section on criticism? —Preceding unsigned comment added by 74.39.212.101 (talk) 04:40, 4 November 2007 (UTC)
What you're describing is a failure of economic theory, not econometrics, to completely describe all economic events. Econometrics is simply a set of statistical tools used to analyze the data. Wikiant 13:40, 4 November 2007 (UTC)
I disagree. Nassim Nicholas Taleb's criticisms in "The Black Swan" and Edward Leamer's criticisms in "Let's Take the Con Out of Econometrics" (http://www.jstor.org/pss/1803924) would be quite appropriately summarized in a criticism section on this article. Lippard (talk) 04:18, 2 June 2008 (UTC)
[edit] Content
Going along the lines of the previous editor, econometrics is not mathematical economics (as falsely claimed by the page), nor is it statistics. Mathematical economics seeks to use deterministic methods and is generally axiomatic in nature. Econometrics tends to be empirical in nature and is by its very nature stochastic (though one could say it accepts the output of mathematical economics as an input, providing a theory-based approach to statistics).
I wanted to add some content to the introduction that differentiates econometricians and economic statisticians but I am unsure where to put it. The main difference is the economic statistician does not perform statistical tests upon the data to verify economic theory, only collects it.
I might go into detail about the 4 main components of econometrics (economic theory, economic statistics, mathematical statistics, and economic theory) how they make up econometrics, and how they are different from econometrics.
- I don't think it matters too much where you put the differences, since it can always be moved. Just choose somewhere that makes sense for you. Expanding the article sounds like a good idea. -- Jrincayc 15:01, 4 Feb 2004 (UTC)
Good points. We also need some discussion of econometric models, what they are, how they are constructed, and how they are used for prediction and policy simulation. Let me think about how to do that. Any suggestions -- F. Gerard Adams (—The preceding unsigned comment was added by 209.6.125.23 (talk) )
The main page should atleast give links to some of the main ideas behind econometrics and regression anaylsis, such as Guass-Markov, Method of Moments, Logit, Probit, Tobit, Standard Errors, and Heteroskedasticity... and they should become articles under econometrics... -- LG (—The preceding unsigned comment was added by Leigao84 (talk • contribs).)
I started an edit with the aim of trying to find a clearer example. I began a list of other methods that is pretty incomplete, but I think that a list would be helpful. EconProf86 01:45, 9 July 2007 (UTC)
[edit] Justification of equation
lnY = − 8.4009 + 0.6731lnX2 + 1.1860lnX3
My question is that -8.4009 it is an intercept or not and why? —The preceding unsigned comment was added by 203.81.192.92 (talk)
It is a log-linear model and ln0,00022466=-8,4009 so it is an intercept.Soulviver 23:29, 30 April 2007 (UTC)
[edit] Er....
I'm sorry, I can't understand a word of this article. Any chance of a simpler introduction? Adam Cuerden talk 02:27, 14 March 2007 (UTC)
[edit] Rewrite
There need to be a major rewrite to the article. I've posted a link under Wikipedia talk:WikiProject Economics for it.Leigao84 22:03, 8 May 2007 (UTC)
[edit] Difference between applied statistics
I took out the line about difference between statistics and econometrics from the intro and the reference to Granger causality.Leigao84 22:03, 8 May 2007 (UTC)
[edit] Econometrics Packages
The list of econometrics packages is a bit ad-hoc -- several of the packages listed here are relatively minor in terms of importance to the field, and others (such as LIMDEP) that have well-established and strong user communities among econometricians are missing. Some of the packages listed here are more popular for teaching than for research, or are not particularly popular among econometricians at all. I didn't edit the list, however, as I don't think my judgment of what should be listed here (and in what order) is necessarily any more objective than anyone else's. My list would definitely include STATA, SAS, LIMDEP, SPSS, TSP, SHAZAM, eVIEWS. I would also include mathematical programming languages that are widely used by econometricians (even if they're not quite the same as the "canned" statistical packages), such as GAUSS, R, S, and MATLAB. Can anyone suggest criteria for this section? 68.148.229.84 21:54, 6 June 2007 (UTC)
I edited the list a bit, and omitted the packages that I didn't believe that econometricians used (in my experience as an econometrician). Some of the packages listed were obsolete. I thought sorting the packages might be helpful as well. I may have editted too deeply, however. EconProf86 01:42, 9 July 2007 (UTC)
TSP has been added. After providing econometric software user support for many years in a 150 person economic research institute, I conclude that TSP has been very well designed and is not obsolete. I agree with the 6 June 2007 posting above that it should be included in the list. Arie ten Cate (talk) 20:44, 10 January 2008 (UTC)
[edit] Simple Example too complicated
It's my belief that a log equation is not simple to a general audience. Why not use the MPC? Burkander 02:04, 31 October 2007 (UTC)
[edit] Logarithms and Utility
A disclaimer: I am not an econometrician. However, I believe the use of logarithms in econometric models has little to do with standard uses of logarithms, i.e. to convert multiplication into addition, and lots to do with the shape of a logarithmic curve, i.e. convex and monotonic. This is the shape needed to describe typical Utility functions, whereby increasing availability of a dependent variable (e.g. income) produces increasing utility to the consumer (e.g. happliness) that are reasonably described by a convex monotonic relationship. In plainer terms (I hope!), an increase in income produce increases in happiness, but such that the rate at which happiness increases reduces for larger incomes. Logartihmic curves look that way, so are a convenient utility function. P.r.newman (talk) 09:21, 3 April 2008 (UTC)
