[R-sig-finance] VAR, VECM, Kalman, ... non-R software recomme ndations?

[Eric Zivot]

I have used Eviews for many years, mostly in my graduate time series class.
I find its VAR functionality to be very good and easy to use. You can do
traditional VARs as well as structural VARs with a push of a button. The
cointegration/VECM functionality is also quite good and easy to use. You can
easily impose restrictions on the VECM parameters, produce forecasts, do
backtesting etc. Eviews excels in dynamic multivariate systems estimation.
You can create very flexible multivariate models and estimate them using
GMM, MLE etc. You can also estimate general models by MLE. The main
drawbacks To Eviews are that it lacks a good programming/scripting language
(it has one but it is not intuitive to use); and (2) you cannot write custom
functions - you are stuck with its canned routines (which are very good).
Note: version 5 of Eviews looks like a very good update, particularly if you
are interested in doing dynamic panel data analysis.  The book on
macroeconometrics by Favaro has many advanced VAR modeling examples using
Eviews. 

I find Eviews excellent for teaching and for doing routine time series
regression etc. In fact, Jeff and I modeled the FinMetrics VAR functionality
to match what EViews does. 

I used RATS many years ago. It is less friendly than Eviews but is more
programmable. It is used by many researchers in economics. Also, Walter
Enders has a RATS companion for his applied time series econometrics book.
If you want to do very fancy VAR analysis then RATS is probably better than
Eviews. In particular, RATS can do a variety of Bayesian VAR models
(FinMetrics also).

-----Original Message-----
From: r-sig-finance-bounces at stat.math.ethz.ch
[mailto:r-sig-finance-bounces at stat.math.ethz.ch] On Behalf Of Dirk
Eddelbuettel
Sent: Tuesday, August 24, 2004 5:30 AM
To: Pfaff, Bernhard
Cc: R-sig-finance at stat.math.ethz.ch
Subject: Re: [R-sig-finance] VAR, VECM, Kalman,... non-R software recomme
ndations?


Thanks to all for the continued feedback.  I am now leaning towards
Rats/Cats. I had used Rats a bit in grad school and have experienced its
idiosyncratic nature which may well drive me up the wall. But it is focussed
on what I need, and half the price of Stata.

One last question: Between Rats and Eviews, would anybody speak in favour of
Eviews?

Dirk

-- 
Those are my principles, and if you don't like them... well, I have others.
                                                -- Groucho Marx

_______________________________________________
R-sig-finance at stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance