Another (the last) one bites the dust

The last course, that is. International Banking and Finance. Was not a great display of econometric skills (deh, that's the value of this literature just like in some strains of business administration) but at least it was a chill, thanks to the instructor: Dr. Degryse from KU Leuven and Tilburg. "Plenty" of time now to get to business and figure out a topic for the next paper and take care of a zillion other administrative stuff.

The next step: Powerpoint Presentation

After working hard on it, it is only natural that you get to present your creation in public. This week is just the ASP conference paper, but a good rehersal for things to come (e.g. the next DIMETIC session in Strasbourg). Anyways, I am (again) really pleased with both the content and the look of my presentation, which is really strange:))..knowing how peaky I am, but good overall.

Finally, I almost forgot. A friendly and funny reminder with what and how NOT to do a presentation. Enjoy!

An outside view of (these cynical) Economists

A nice article about it in NY Times

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.

Updating...some older paper

I have started to update my previous paper analyzing patenting activities and determinants of national innovation systems in the transitional Eastern European space. Well, as always..the good the bad and the ugly features and issues are there. After all it is not like writing the whole paper again...but it takes a lot of time (too much) as well. So...I keep my fingers crossed

Forced Break from my Research

This week I cannot work almost anything since I have to attend and pass the course in Monetary Policy by Mr. Eichenbaum from NorthWestern, which is a funny/smart/etc teachers...but there is a problem: these huge macro models especially monetary ones are not my thing. at all. so, hopefully it will not be too painful.

Eviews 6 - Small is Beautiful

When I first ran anything on Eviews this Autumn I was quite skeptic about its true power. But when you think about it, this program knows an amazing amount of stuff for its size. And now with version 6 it gets even better (than Stata? - in some issues yes).

Not so small after its latest makeover it features some amazing new things, some of which are not available in Stata not even via the ssc channel.
See for yourself.

The things that got me jabbering was panel cointegration tests: of Pedroni (1999) up till now was the RATS's turf or Kao (2000) requiring Matlab or Gauss platforms.

Seems I was right about Stata10

Unfortunately, it looks like indeed Stata10 has some serious bugs. The thing that I hate the most is running a do file with multiple specs regressions and get an error in the middle of the program because one of the variables has magically dissapeared. Sometimes is a transformation var only which can be re-done manually from the command line, but when the base variable is gone...not much one can do but clear and reload the data. Perhaps a 9.2 downgrade is not that bad after all.
Other people had the same problems.

It takes a Genius to Read (some) blogs

At least, it seems so. More details here.

Adding control variables

The stock of domestic R&D is a hard to get and valuable variable in the case of developing nations. On top of that I have to collect some data, measures of human capital…of which I opted for the world wide known Barro and Lee 2000 dataset and perhaps some other schooling stuff from UIS Unesco and the WID 20007.
I already used the BL 2000 and the tertiary enrollment and they seem to impact significantly the TFP spillovers, which is good. :)

Finishing the business

YES, I am extremely happy to say that I have finished the FDI spillovers…sometime Sunday night before the meeting with my advisors. It wasn’t a must after all but the numbers proved to be significant despite their inherent deficiencies. However, I have to look closer at some countries that are not represented anymore in my FDI restricted sample…but again, how much and who is investing in Uzbekistan or Tajikistan? So far I don’t know but it’s on my wish list to cross-check it with the UNCTAD data.

FDI related spillovers - The hard way

There aren’t many choices when you want to have flows and stocks of FDI (bilateral) among many countries. Usually people go for the OECD International Investment data which to my humble opinion is quite lame. I mean, I would expect at least to have flows among all OECD countries starting the 1990s but I guess it is to much to ask. So, just imagine how many observations can you get for, Armenia, par example. Well, usually 10-20 out of the 25 OECD times 16 years…well out of 400 let’s say. Is that workable? 
So, I decided to go with UNCTAD and their statistics. Not so inspired here either. Why? Well, they have it all at the aggregated level (total flows, stocks) but bilaterally it’s a bit better than OECD but the coverage and breakdown for developing countries is even worse (up to the national statistics office who collect the data). Moreover, the time varies 1990-2003, 04 or 05 and NATURALLY, the numbers do not match OECD ones .
The cherry on top: they are only available in PDF form!!! Not even excel… which is just great. Now I have to spend at least half an hour for each country migrating the data from PDF to Excel (which is not that easy in itself) then clean it, structure it and finally compute flows, shares, and fdi related spillovers. I just hope at the end of the day it wasn’t ALL in vain…since the data is quite scarce.

The never ending story

Naturally, the flows of FDI even from the OECD nations among themselves is highly “volatile”, quality wise with a decent amount of missing observations. Not to go deeper into the problem when you look at the Eastern European ones. After computing all these stocks and flows of capital, R&D or trade I really do not feel like spending my entire life searching for bits, pieces and approximations of all the incoming FDI to my 27 transitional countries in the sample. So I guess I will probably stick (in the 1st phase at least) to trade related spillovers, since, I will also probably have to invest some time and energy into econometrics…testing for unit roots and applying panel co-integration techniques.

Finished part 1 and 2

I just finished part 2, let’s say of my paper, which is computing the trade related spillovers from 25 OECD countries towards another 20 Western plus 27 Eastern and Central Asian (former Soviet Union) states using bilateral trade data from DOTS 2007 from IMF. The data looks good, it would be marvelous if one could get the same accuracy with FDI data but that’s definitely not the case . The struggle begins now…

November 19, 2007

Doing all that spreadsheet tedious work

Today I continued to process all the import-export data for the OECD countries and I reached Mexico. Ten more countries to go. I have 50 receipients some of which are a bit problematic since one needs to aggregate (Belgium-Luxembourg union) or take out countries like former Czechoslovakia, URSS and very new entities without any values (Serbia on its own, Montenegro, Kosovo, etc).