V krizi smisla tiči misel



Nekaj ekonometričnih videov

Nekaj videov s področja ekonometrije, najdete jih tudi tukaj, urejeni so po letih.

Econometrics of Matching (Econometrics Journal Special Session)

Chaired by Jaap Abbring, Tilburg University

Estimating Transfer Frictions in the Marriage Market:
By Alfred Galichon; Sciences Po
Presented by: Alfred Galichon, Sciences Po

Multidimensional Skills, Sorting and Human Dimension Calculation
By Jeremy Lise, UCL and IFS and
Fabien Postel-Vinay, UCL, IFS and SciencePo
Presented by Jeremy Lise. UCL and IFS

Large Dimensional Models

Monday 7th April 2014 Including: Dynamic Factors and Volatilities in Panel Data by Matteo Barigozzi, London School of Economics and Marc Hallin, ECARES, Universite Libre de Bruxelles and ORFE, Princeton University and Large Panel Test of Factor Pricing Models by Jianqing Fan, Princeton University with Yuan Liao and Jiawei Yao.


Chaired by Richard Smith, University of Cambridge
Speakers: Yuichi Kitamura, Yale and Stephane Bonhomme, CEMFI

Discussant: Lars Nesheim, UCL

(RES 2012 Annual Conference Cambridge – Special Session=

Econometrics of Forecasting

Raffaella Giacomini (UCL) : Economic Theory and Forecasting
Slide Pages: 1 – 18
Timing: 49sec – 35min

Siem Jan Koopman (Free University Amsterdam) : Likelihood-Based Dynamic Factor Analysis for Measurement and Forecasting
Slide Pages: 19 – 24
Timing: 35min 37sec – 73min

Brendan McCabe (University of Liverpool) : Discussion
Slide Pages: 24 – 25
Timing: 73min 14sec – 84min 37sec

(RES 2011 Annual Conference Royal Holloway – Special Session)

Nonparametric Identification: Current Issues and Problems

Rosa L Matzkin (UCLA) : Identification in nonseparable models using shape restrictions
Slides: 35 – 155
Timing: 14min – 47min

Andrew Chesher (UCL) : Identification with multidimensional heterogeneity
Slides: 119 – 183
Timing: 48min – 1h 26min

Victor Chernozhukov (MIT) : Discussion
Slides: 186 – 202
Timing: 1h 27min – 1h 40min

  • Share/Bookmark


About “mathiness” (in English)

Zapisano pod: Ekonomija — andee - 1.06.2015
Tagi: ,

Paul Romer raises important issues in his widely debated »Mathiness« article: http://paulromer.net/wp-content/uploads/2015/05/Mathiness.pdf. But it is important not to throw away the baby with the dirty water. My experience with economics (and most of today’s social science in general) is that its today’s language is indeed in a kind of lemons market equilibrium: with no »fancy« math model or econometric method nobody will listen to you today. Which is bad – there is no real need for a real development of mathematical and econometrical models and significant new findings, the most important thing is to publish your article in a »fancy« journal, no matter what it takes. Publish or perrish, commodify or die.

But there is another extreme which Romer nicely warns against: »After all, how would Piketty and Zucman have organized their look at history without access to the abstraction we know as capital? Where would we be now if Robert Solow’s math had been swamped by Joan Robinson’s mathiness?« There is no »mathiness« in math and my only opinion on this is nicely summarised by this video:

YouTube slika preogleda

The problem of contemporary economics is not overmathematisation, it is undermathematisation, however strange this may sound to some. Most models I have dealt so far in my research (and I’ve done some modeling) were at the end limited to some rather convenient optimisation problem, solved by Lagrangian, Bellman, Hamiltonian or other functions. More or less, always the same. There are great possibilities of building more complex and, perhaps, more accurate and predictive economic models using math. And probably not only economic models, but also models in social science in general.

Sincererly said, as someone who has done some slight work also in social theory, even philosophy, most of the theoretical articles I listen to lately often appear to me pretty dull and trivial, with an impression I could write tones of these if I just wanted to be published. I realise I might have a bias regarding this, but going really deep one has to understand the »math« (not mathiness), the structure, the relationships behind the phenomena he observes. Often this is not visible by naked eye, by pure speculation, so one needs some tools to do this, if he is really interested in »going behind« (maybe not a good word, but it pretty much describes what I want to say). It is possible he will make mistakes, but if he doesn’t do this, is this not bluffing? Just pretending everything harder to understand is just some »neoliberal« nonsense, as some leftists would say. Well, it probably depends on the individual decision, but it is hard to accept people only talking about »bullshit« in something, not done any serious, empirical work to really get to what is »good« there.

So, in the end – as said, there is no »mathiness« in math. Still, it remains that there is only »good work« and »bad work«, people who want to go deeper and people who only want to make their money, publish something and go back to bed. And, it is one’s decision on which path to choose.

  • Share/Bookmark

Blog V krizi smisla tiči misel | Zagotavlja SiOL | O Sistemu |