André Berchtold

Softwares

MARCH

New version 1.1

MARCH is a free software for the computation of Markovian models for homogeneous and non-homogeneous data sequences. The standard distribution comes as a set of Matlab functions requiring Matlab 5 or higher on any computing platform. In addition, a compiled version running independently from Matlab on the PC Windows platform (95 or higher) is also available. See the user's guide for more details.

You can download the following files:

The user's guide in Postcript format. March 27, 2001.
The user's guide in Pdf format. March 27, 2001.
The standard distribution. Version 1.1, March 27, 2001.
The compiled version of MARCH (PC Windows only). Version 1.1, March 27, 2001.
The files used in the first example of the user's guide. March 27, 2001.
The files used in the second example of the user's guide. March 27, 2001.

In addition to the user's guide, the two following papers give a lot of details about the Markovian models computed by MARCH:

Berchtold, A. (1999) The Double Chain Markov Model. Communications in Statistics: Theory and Methods, 28 (11), 2569-2589. Preprint: Technical Report 348, Department of Statistics, University of Washington.
Berchtold, A. (2002) High-Order Extensions of the Double Chain Markov Model. Stochastic Models, 18 (2), 193-227, 2002. Preprint: Technical Report 356, Department of Statistics, University of Washington.

Copyright: The rights of MARCH belong to André Berchtold. The software can be freely copied, distributed and used, but it cannot be modified or sold without authorization.

GMTD

GMTD is a software for the computation of the Mixture Transition Distribution (MTD) model for high-order Markov chains. It comes as a set of Matlab functions running under Matlab 5 or higher on any computing platform. In addition, a compiled version running independently from Matlab on the PC Windows (95 or higher) platform is also available. See the user's guide for more details.

You can download the following files:

The user's guide in Postcript format April 23, 1999.
The set of Matlab functions. Version 1.0, April 23, 1999.
The compiled version of GMTD (PC Windows only). Version 1.01, August 31, 2000.

The MTD model was introduced first in

Raftery, A. (1985) A Model for High-order Markov Chains, JRSS B, 47 (3), 528-539.

A complete description of the model and of the optimization method used in GMTD is given in

Berchtold, A. (2001) Estimation in the Mixture Transition Distribution model. Journal of Time Series Analysis, 22 (4), 379-397. Preprint: Department of Statistics, University of Washington, Technical Report 352 (Postcript file)

Another good reference is

Berchtold, A., Raftery, A. (1999) The Mixture Transition Distribution (MTD) Model for High-Order Markov Chains and Non-Gaussian Time Series. Forthcoming in Statistical Science. Preprint: Technical Report 360, Department of Statistics, University of Washington.

Copyright: The rights of GMTD belong to André Berchtold. The software can be freely copied, distributed and used, but it cannot be modified or sold without authorization.

Last modified: October 10, 2002. Webmaster