Tuesday, January 31, 2006

The model got it right after all

Last year I had this post: What the Eagle Saw. The Ateneo Macroeconomic Forecasting Model (AMFM) generated a forecast of 5.3% growth for the Philippines in 2005. This set the upper bound for the forecasting range; the lower bound (4.5%) was from an alternative model, plus adjustment for a pessimistic scenario due to political uncertainty.

Well the model was more or less on track: actual growth was 5.1%. The economy bucked political concerns to outperform most other forecasts (i.e. the competition!)

As the modeler I can't deny the ego boost. To be fair though, one needs to examine AMFM forecasting performance over several periods. One hit does not a reliable model make. On a consultancy basis I'm working on the AMFM to improve its forecasting performance.

Getting to the point: Ateneo will be holding another Eagle Watch economic briefing this coming Wednesday, February 8. Forecasts for this year will be unveiled. Hopefully we can repeat our forecasting performance for 2006.

Off for another trip, this time to Samar in central Philippines (Visayas). No, not related at all to macroeconomic forecasting! Next post at the end of the week.

8 comments:

Rizalist said...

I love econometric models! Maybe we'll run into each other at the Feb 8 meeting. am an old friend of Ciel Habito's he usually attends I guess, though this would be my first time in a long time.

Econblogger said...

Hi DJB,

Ciel is actually my immediate boss for this project. (We go some way back too, he was my undergraduate thesis adviser in UPLB). He is the main presenter for the economics component (there is also I believe a political outlook given by the Poli Sci Department of ADMU). I'll be at the sidelines just in case somebody who loves econometric models gets a bit curious. Hope you can make it!

Amadeo said...

I understand the elation because the model bucked the lower forecasts of the "competition".

Yet, I read this from Malaya:
http://www.malaya.com.ph/dec08/edit.htm

and I'm confused. The Ateneo group appears at odds with the government's position.

BTW, our family is close and related to Romy Neri.

Econblogger said...

Amadeo,

Yep, that's why we gave an interval forecast (most do). Our interval though was higher than most of the others. Ciel's assessment is certainly critical of the government spin on the economy, sure, and I agree with most of it. However we realize that, despite all those negatives, the economy could remain resilient. It doesn't mean though that government is doing the right thing - growth could be higher, more sustained, and pro-poor, if better policies were in place.

cvj said...

For consistency's sake, i would like to ask how your view of the AMFM forecast squares with your 04-Jan,2006 post on 'How to Make a Forecast'. Is there such a thing as good forecasts (e.g. AMFM) and bad/ill-advised forecasts (e.g. exchange rates) or is it just a matter of getting the predictions right?

Amadeo said...

cvj:

In a light-hearted way of saying some thing serious, I say trust the forecaster who says that his forecast is not a sure thing.

Because one who says he knows and has factored in all the possible variables, realizes it is impossible to make a very accurate forecast.

As this economist here says, play the martingale:

http://www.econbrowser.com/archives/2006/01/when_is_it_good.html

cvj said...

amadeo, thanks for the link.

Without Borders said...

have you noticed that multilaterals like the ADB and the WB always have very low forecast re Phils? And they are always eating crow!