Endogenous growth course

This week the reason for taking on this ASP program will reveal itself to me. Innovation and Growth (taught by Dr. Zilibotti of University of Zurich)is basically what made me choose this program which overall it has a very good one anyways.

Akward Mobile office

For the last hours in the Budapest airport, I had to beat down the fatigue through work. It's a funny feeling though to work while hundreds of people pass you by. Loud music helps as always. Some future ideas and conference applications have been successfully born in the time spent here.

Finishing touch

Yes, I finally called it off. After a couple of robustness checks and sensitivity analysis, which in the end I have decided to not report even, due to the lack of space, I decided that is enough. Thus, I did a reading proof and attached the JEL codes to my abstract and my brand new paper is done. The result resembles me much like a Huf house, meaning exquisite, clean and robust, so I am quite excited about it and I will probably submit it somewhere before this summer.

Tweaks & tests on my first essay - revisions

The Breusch-Pagan Lagrange multiplier test for random effects tests if variances of groups are zero. The BP LM values reject the null only in the first and second model, suggesting that random effects should be used in these cases against an OLS estimator.
Using a likelihood-ratio test, homoskedasticity is rejected firmly in all models. Thus I take this into account by using White cross-sectional standard errors which are robust. Beyond this, serial correlation could be biasing the estimates so I use the test described by Wooldridge (2002) and implemented in Stata. However, the null of no serial correlation can be rejected at 5 percent significance lavels in all relevant models, proving that this is not an issue for these estimations.
To make sure that my regressions are not spurious, I perform the most common two panel unit root tests involving regressions on lagged difference: Levin Lin and Chu (2002) which assumes a cross-sectional common unit root and Im, Pesaran and Shin (2003) that allow for individual unit root processes across sections.

Satisfaction of a good job

I have just finished writing my paper last Friday. After a dog-week period when I worked around the clock to write it and do all the necessary analysis, the result looks promising and the paper very professional. In the end, I included various cointergration tests (e.g. Pedroni and Kao) plus all the relevant panel unit-root tests (Levin Liu and Chu, Breitung, Im Peshran and Shin, Breitung and Hadri). Eviews gives all that almost for free. Naturally, nobody knows the true spec so one includes everything (trend, itercept etc) and hopes for the best (I(1) that is). Happily for me it worked like a charm. On top of that, besides the usual OLS fixed effects model, I also managed to get some DOLS estimates which seem to outperform both OLS and FMOLS in small samples. So , everything looks good and healthy around here. Just need to finish the fine tunings and additions and send it to some journal.:)

Writing is the name of the game

I have started to write up the paper yesterday. Although I have bits and pieces already, it is not that easy to put them all together, and I would really like to avoid duplicative efforts. On the other hand, it's just a first draft (from not more, hopefully than two or three) so perfection should come second.