MAXIMUM LIKELIHOOD MT 1.0 (MLMT)

Maximum Likelihood MT 1.0 (MLMT) is a new product from Aptech Systems for use with the GAUSS Mathematical and Statistical System(TM) or GAUSS Engine(TM) programs. Using the same structures as Constrained Maximum Likelihood MT, MLMT has the ability to compute some of the derivatives analytically, leaving the others to be computed numerically.  It is also able to place bounds on the parameters.

New Features:

• Simple bounds may be placed on parameters; bounds are all that are needed for many models, and MLMT is faster than other applications with more general types of constraints
• Structures, in particular DS structures for handling data, and PV structures for handling parameters
• New numerical derivatives, user provided analytical derivatives can compute a subset of the derivatives, the rest will be computed numerically MLMT uses standard Newton and quasi-Newton descent methods and it allows for only one kind of constraint, bounds on the parameters

Major Features of Maximum Likelihood MT

• Structures
• Simple bounds
• Hypothesis testing for models with bounded parameters
• Log-likelihood function
• Algorithm
• Secant algorithms
• Line search methods
• Weighted maximum likelihood
• Active and inactive parameters
• Bounds

Maximum Likelihood MT 1.0 requires the GAUSS Mathematical & Statistical System v8.0 or GAUSS Engine v8.0 and is available for Windows/LINUX/UNIX.