TROLL is an integrated software system for econometric modeling and statistical analysis. TROLL is the primary econometric modeling, simulation and analysis tool at many of the world's leading central banks, national and international governmental agencies, commercial institutions and research universities. TROLL is designed for both model builders and model users. It includes a database engine for the efficient storage and retrieval of data, powerful analytical tools for the manipulation, estimation and simulation of systems of equations, and a macro programming language and external interfaces for the automation of complex tasks. TROLL offers a complete environment specialized in processing of analytics, including data storage and manipulation, model editing, regression tasks, simulation of models, analysis of properties of systems of equations, design of reports, and interaction with spreadsheets and RDBMS. Underlying the TROLL system is a powerful symbolic differentiator for exceptionally fast and robust performance on nonlinear problems. TROLL's performance, combined with its ability to process models or equations of unlimited size, makes TROLL especially well-suited for solving large and complex econometric and general equilibrium models. TROLL offers a wide array of ready-to-use analytical modules, including interactive calculations with over 200 functions, Census X11 seasonal adjustments, and advanced and efficient estimation and simulation tasks for both linear and nonlinear equations. Capabilities range from simple OLS, GLS and 2SLS methods to nonlinear system estimation techniques such as FIML and leading-edge forward-looking "rational expectations" simulation algorithms. TROLL is available on many different platforms including PCs and various Unix workstations. TROLL has been developed by the MIT and Intex. For detailed information about TROLL, please visit: www.hendyplan.com and www.intex.com |