Growth diagnostics: the dangers of agnosticism

Why does income grow faster in some countries than others? A common empirical approach in recent growth analysis has been to adopt an 'agnostic' approach and let the data do the talking (i.e. weak priors). But a new paper by Antonio Ciccone from Barcelona's ICREA-Universitat Pompeu Fabra, and ECB economist Marek Jarocinski, question this approach. In their new ECB working paper no. 852, Determinants of economic growth: will data tell? (PDF), they show that small differences in the comparative income data can have a substantial effect on the outcomes:

As many potential explanatory variables have been suggested, these agnostic empirical approaches inevitably need to start out with a long list of variables. We show that, as a result, the growth determinants emerging from these approaches turn out to be sensitive to seemingly minor variations in international income estimates across datasets. This is because strong conclusions are drawn from small differences in the R2 of different growth regressions. Small changes in the relative fit of different models—due to Penn World Table income data revisions or methodological differences between the PWT and the World Bank income data for example—can therefore lead to substantial changes regarding growth determinants.

Their analysis clearly shows that agnostic growth regressions can be sensitive even to small data revisions. They suggest that "the available income data may be too imperfect for agnostic empirical analysis". So what to do? Stronger (and fewer) priors:

At the same time, we find that the sensitivity of growth determinants to income differences across data revisions and datasets falls considerably when priors regarding potential growth determinants become stronger. That is, the data appears good enough to differentiate among a limited number of hypotheses. Empirical models of the typical size in the literature, for example, tend to point to the same growth determinants using different versions of the PWT or the World Bank income data.

Researchers who want to continue giving equal a priori weight to all potential growth determinants in the literature, should consider shrinkage priors, explicitly incorporating priors about measurement error in the income data, or implementing Zellner’s (2002) adjustment for data quality.

Comments

[ZKu]

I though the approach they criticize goes under the heading of "Barro's everything but the kitchen sink" regressions. The Growth diagnostics is label I would associate with Dani Rodrik's work, that is much more interesting than [largely discredited] cross-country regressions.

http://rodrik.typepad.com/dani_rodriks_weblog/2007/11/doing-growth-di.html

[Glenn Athey]

Its just the same as all the stuff I've looked into about the accuracy of economic forecasting. In models such as Cambridge Econometrics/ IER At least 50% of the forecasting error is usually due to errors in source data.

The predictive strength of modelling anything is only as good as the source data used to build the model.

[ilanlar]

"I though the approach they criticize goes under the heading of "Barro's everything but the kitchen sink" regressions."

yes i think too.

ilan

[oyun]

Thanks, this information it was very useful to me.
Oyun

[Oyun]

very thanks you

[Oyun]

very thanx for you

[oyun gemisi]

Congrulations for this nice web design..Usefull..

[oyun]

oyun
oyun
oyun
Thanks, for this information and news it was very useful to me

[eleman]

harika bir olay very thanx for you..!Oyun
Oyun

[cicicocuk]

Thanks so much.
Oyun

[oyun]

Its just the same as all the stuff I've looked into about the accuracy of economic forecasting. In models such as Cambridge Econometrics/ IER At least 50% of the forecasting error is usually due to errors in source data.

[Estetik]

While user-generated content is not new, what is new is how easy it is to publish and manage. In 2005 Sun CEO Jonathan Schwartz wrote in his blog: “Simplicity changes the world. Convenience is a force multiplier.

[Estetik]

For those of you thinking that if they implement this it will eliminate some of the waiting and lines…

[Estetik]

Thanks for this nice blog...

[Maç Oyunları]

have followed your writing for a long time.really you have given very successful information.
In spite of my english trouale,I am trying to read and understand your writing.
And ı am following frequently.I hope that you will be with us together with much more scharings.
I hope that your success will go on.

[alışveriş]

As many potential explanatory variables have been suggested, these agnostic empirical approaches inevitably need to start out with a long list of variables. We show that, as a result, the growth determinants emerging from these approaches turn out to be sensitive to seemingly minor variations in international income estimates across datasets. This is because strong conclusions are drawn from small differences in the R2 of different growth regressions. Small changes in the relative fit of different models—due to Penn World Table income data revisions or methodological differences between the PWT and the World Bank income data for example—can therefore lead to substantial changes regarding growth determinants.

[okey]

thanks

[sikiş]

thanks

[min]

Most of us have had an experience which reminds us the world is getting smaller. Perhaps when you went to some place far away from the city you live in and thought you knew nobody there, you were surprised [URL=http://www.bodalliance.com/ghd-wholesale-1.html]ghd[/URL]
[URL=http://www.disnet1.org/]wholesale[/URL][URL=http://www.disnet1.org/]wholesale electronics[/URL][URL=http://www.uggbootsshoes.com]ugg boots[/URL]
[URL=http://www.uggbootsshoes.com/]wholesale ugg boots[/URL][URL=http://www.disnet1.org]Wholesale handbags[/URL][URL=http://www.disnet1.org/]wholesale clothing[/URL]
[URL=http://www.disnet1.org/]Wholesale jewelry[/URL][URL=http://www.bodalliance.com]wholesale clothing[/URL][URL=http://www.bodalliance.com/]wholesale handbags[/URL]to find that you ran into one of your old classmates on the street! Then both of you cried out, “What a small world!” Why is the world getting smaller and smaller? For one thing, modern technology has created various means of transportation. And they are more and more available. Buses, trains, planes and [URL=http://www.neromn.com]wholesale furniture[/URL][URL=http://www.neromn.com]Furniture Wholesale[/URL][URL=http://www.askthecru.com]Wholesale jewelry[/URL][URL=http://www.harstiner.com]Ceramic tile[/URL]
[URL=http://www.microsd-cards.com]Micro sd cards[/URL][URL=http://www.pupbun.com]Wholesale clothing[/URL][URL=http://www.ememe.net]Wholesale Jewelry[/URL][URL=http://www.ememe.net/]Wholesale fashion jewelry[/URL][URL=http://www.ememe.net/]Wholesale costume jewelry[/URL]
[URL=http://www.ememe.net/]ugg boots[/URL]taxis are found in most cities of the world. They carry goods and passengers to and from every corner of the world. For another thing, with the development of modern society, people are in contact with one another more frequently. Everyday we must deal with a lot of people unknown to us before. The more advanced a society, the more contacts we make with others

[min]

Knowledge is power, ghdwholesaleespecially scientific and technological knowledge. wholesale electronicsScience and technology are the motive power of the social development. Without them human society could never have developed from primitive society to modern society. Therefore, to conquer anugg bootsd transform nature, we must masterwholesale ugg bootsWholesale handbagswholesale clothingWholesale jewelry scientific knowledge. However, social knowledge is also essential. wholesale clothingwholesale handbagswholesale furnitureFurniture WholesaleWholesale jewelryWithout it we can not understand society and don＇t know the law of the social development. As a result we are unable to tovern society. Therefore, besides scientific knowledge we need to master social science, philosophy, politics, history, aesthetics, etc, so that we can know society from all perspectives and form a correct world outlook.To meetCeramic tileMicro sd cards
Wholesale clothingWholesale JewelryWholesale fashion jewelryWholesale costume jewelry
ugg boots new challenges in he 21st century, we university students should lose no time to acquire as much knowledge as possible so that we will become qualified successors of the socialist cause.

[Metehan]

Thanks for this nice blog...

[dedektif]

thanks

[sinema izle]

great post 10x

[porno]

thanks you

[neon tabela]

expression and clarity of the information very clearly. injury topics. Thank s for my analysis. quality site.

[xxx movies]

expression and clarity of the information very clearly. injury topics. Thank s for my analysis. quality site.

[lise]

Great post...

[Domuz Gribi]

good information.

[Domuz Gribi]

Thanks Man...

[sikiş]

very nice site
Thank you for sharing

[Okey]

thanks for you..

[yaşam koçu]

thanks a lot

[koçluk]

good news. thanks for the sharing.

[life coaching]

economy is one of the guns of countries. and i have found most important informations at your page.

[trainings]

very useful article thanks

[William Right]

I have been following your website and your analysis. This one is excellent! www.bigjobsboard.com

[saç ekimi]

I'm really very useful to follow a long-time see this as a blog here Thank you for your valuable information

[saç ekimi]

Very interesting article with unique writing style. Please do more this kind of post.

[domuz gribi]

great post 10x

[rent a car]

Great job, keep it up. thanks