Version 5.0, Copyright (C) 2010
Fernando V. Lima, Murali R. Rajamani, James B. Rawlings and John W. Eaton. Modified for modern distributions by Aaron Webster.
ALS is a free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 3, or (at your option) any later version.
ALS is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the text file LICENSE for more details.
The latest released version of ALS is available via the Rawlings Group website.
To compile ALS, you will need a recent version of GNU Octave. You will also need the Sundials toolbox (version 2.4) to compile the linear time-varying ALS (LTV-ALS) module. The linear ALS module is also compatible with recent versions of MATLAB (tested under version 7.6, R2008a).
Questions, comments, bug reports, and contributions should be sent to the [email protected] mailing list.
The linear ALS technique is implemented using the "als_sdp_mrQ.m" function. Files "simulate_data8.m" and "motivating example.m" are examples of ALS implementation for simple systems. Specifically, please see the last few lines of "simulate_data8.m" for a sample implementation of the ALS method. The diagonal ALS technique, to estimate only the diagonal elements of the covariance matrices, is implemented using "als_diag.m", and "simulate_data8_diag.m" is an example of its implementation. The linear time-varying ALS technique for nonlinear and time-varying models is implemented using the "ltv_als.m" function and file "case1_als.m" provides an example of its implementation. The rest of the functions are supporting functions that are not directly called. Documentation for each function is also available by typing "help function_name" in the octave command window.