A queestion: what are major statistical macro models?

what kind of model are you looking for?check out DSGE models

QuoteOriginally posted by: JamesH83what kind of model are you looking for?check out DSGE modelsI'm looking for models associated with macroeconomics and finance. I am also more interested in monetary and FX policies.If you know some good references, please do not hesitate to let me know.

any one, pls???

There is a quiet new literature linking macro factors estimated from large panels with excess returns and volatility. Try: www.econ.nyu.edu/user/ludvigsons/

BestServedCold,What is panel data analysis?

Econometric methods tend to be either Time series models (under ARIMA, Cointegaration method) or Cross Sectional (Latent variable, Logit...regression method).You are either studying a variable across time or you are studying a person/country/? behaviour across several variables.Panel methods allow you to study both time series and cross sectional albeit controlling for certain factors. The most common applictaion of the panel method is in controlling for the fixed effects or random effects. Say we are measuring teh determinants of economic growth across countries. We have GDP, int.rates, M2 growth, etc, across 50 years. The fixed effects and random effects models are aimed to capture the country specific effects via a a fixed/random effects coeffcient, so you can distinctly seperate the country specific idiosyncracies. I'm a bit rusty but if I recall, part of the purpose of teh random effects was to capture within group and between group variation.

As for macro modelsStructural based models tend to be a whole system of equations based on various economic relationships. These systems would tie in the monetary sector, real sector, etc, etc in a whole system of equations. Just think of it as multiple regressions run simultaneously. Some calibrations is required to allow the iteration to complete. Usually done as a result of a fudged constant/variable.Structural variables while theoretically sound - take way too long to calibrate correctly - and by extension run.Besides changes economic strcuture tend to mess up the modelers system.Othere structural methods include Computable general Equilibrium (CGE) methods. Analytical based method. as opposed to the system of equiations which is more regression based and hence by extension 'statistical'Reduced form equations tend to be more commonly used. less theoretically sound but hey if the objective is prediction - as long as the basic structure is sound . Usually is OK. If you are looking at monetary - chechk out the Monetary Conditions Index and Taylor RuleFor FX, especially ones which are a managed peg - we used to have this thing called the Central Bank reaction function.All this complexity sounds cool. but honestly the simpler methods were usually as good as the most complicated methods for forecasting purposes. And a hell lot faster to get up.

KL,For these two topics, would you please give me a good reference?

Global Insight has structural models for most of the major industrial countries. You have be a subscriber to see how they work though. DRI and WEFA combined into Global Insight a few years ago, so there's really only one major structural econometrics firm left. All the other macroeconomic models seem to be in the universities or the central banks or finance departments....which are often built by Global Insight. They are not much use for capital markets data though, less than useful IMHO. ------------------QuoteOriginally posted by: JA queestion: what are major statistical macro models?

KTE, there is also Oxford Economic Forecasting (www.oef.com). Their 'global model' is a easy to use menu driven application in which you can simulate different macroeconomic scenarios. Another firm that uses large structural models is FERI Research.