I know there is a neural nets option in RATS, but are there any procedures for Multivariate Adaptive Regression Splines (MARS)
or Hinging Hyperplanes (HHP) which have become reasonable alternatives?
MARS was introduced by Friedman (1991) Annals of Statistics, vol 19, pp 1-141.
HHP was introduced by Breiman (1993) IEEE Transactions on Information Theory, vol 39, pp 999-1013.
Both can be represented as tree structures similar to neural nets, but their basis functions are piecewise linear,
affording greater ease of interpretation in econometric applications. Have these been implemented in RATS or
are there plans to do so? (I am familiar with the Bai-Perron procedure in RATS, but it is limited in scope.)
Nonlinear Models
Use this for suggestions about improvements in RATS
-
Ken-Cogger
- Posts: 13
- Joined: Wed Feb 15, 2012 7:58 pm
Jump to
- News and Announcements
- ↳ Forum Announcements
- ↳ RATS Product Announcements
- ↳ Newsletters
- Courses and Seminars
- ↳ Course Announcements
- ↳ ARCH GARCH and Volatility
- ↳ RATS Programming Manual
- Procedures, Examples, and Sample Code
- ↳ RATS Procedures
- ↳ Examples and Sample Code
- ↳ Looking for Code?
- General RATS Discussions
- ↳ Data: Reading, Writing, Transforming
- ↳ Graphics, Reports, and Other Output
- ↳ Help With Programming
- ↳ Other RATS Usage Questions
- ↳ RATS for Teachers & Students
- ↳ Tips and Tricks
- ↳ Suggestion Box
- Econometrics Issues
- ↳ VARs (Vector Autoregression Models)
- ↳ ARCH and GARCH Models
- ↳ State Space Models/DSGE
- ↳ Structural Breaks and Switching Models
- ↳ Panel Data
- ↳ Other Time Series Analysis
- ↳ General Econometrics
- CATS (Cointegration Analysis)
- ↳ CATS News and Announcements
- ↳ CATS Questions