Erwin Kalvelagen

    Erwin Kalvelagen
    GAMS Development Corp.
    1217 Potomac Street NW
    Washington DC 20007
    erwin@gams.com

     
     

    Software

    http://www.gams.com/~erwin/interface/interface.html
    http://www.gams.com/~erwin/interface/interface.pdf
    		 This document illustrates how GAMS can work with Excel, Access, and relational databases such as Oracle, DB2, MySQL and SQL Server. It also covers how GAMS can be called from other progams and environments. Utilities for reading .XLS files, and Access and relational databases are available from here.
    Tools:
    XLS2GMS for reading MS Excel XLS files
    MDB2GMS for reading MS Access MDB files
    SQL2GMS for reading data from SQL databases (SQL Server, Oracle, DB2 etc.); includes support for ODBC
    GDXVIEWER for viewing and exporting from GDX files
    ASK for building interactive GAMS applications
    GDX2ACCESS dumps a GDX file to an MS Access database
    GDX2XLS dumps a GDX file to an Excel spreadsheet
    for download instructions see http://www.gams.com/~erwin/interface/interface.html section 1.31 WTOOLS.
    http://www.gams.com/~erwin/fitpack.pdf Description of GAMS/FITPACK, a package to perform spline curve fitting and interpolation with GAMS. Splines can be used to create smooth functions. In some cases models using splines outperform piecewise linear formulations by a large margin.
    http://www.gams.com/~erwin/gamsqueue.pdf Description of GAMS/ORAQUEUE, a queueing package to submit GAMS jobs in an Oracle environment. A web based submission tool is used to submit GAMS jobs using a standard browser. The code can be used to develop robust Web based optimization applications.
    http://www.gams.com/~erwin/rgams/rgams.pdf Remote GAMS execution from the IDE: use the IDE from your PC to run jobs on a remote UNIX/Linux machine. Supports FTP, Telnet, scp and SSH1/2 protocols.
    http://www.gams.com/~erwin/specfun.pdf New special functions GAMMA(.), BETA(.), LOGGAMMA(.), LOGBETA(.), GAMMAREG(.), BETAREG(.), BINOMIAL(.).
    http://www.gams.com/~erwin/ShellExecute.pdf Execute a program based on the document type.
    http://www.gams.com/~erwin/interface/wlog.pdf Log progress by sending messages to a window (NT/XP only).
    http://amsterdamoptimization.com/pdf/regression.pdf Regression solver. This is a GAMS solver that can solve linear least squares problems and report a suite of standard regression statistics.
    http://amsterdamoptimization.com/pdf/nlregression.pdf Nonlinear regression solver. This is a GAMS solver that can solve nonlinear least squares problems and report a suite of standard regression statistics.

    GAMS Solver Documents

    http://www.gams.com/~erwin/dicopt.pdf GAMS/DICOPT, an MINLP solver. This is a user's guide.
    http://www.gams.com/~erwin/minos.pdf GAMS/MINOS, an NLP solver. This is the user's guide.
    http://www.gams.com/~erwin/snopt.pdf GAMS/SNOPT, an NLP solver based on a large scale SQP algorithm.
    http://amsterdamoptimization.com/pdf/regression.pdf LS, a solver for linear regression (OLS) problems .
    http://amsterdamoptimization.com/pdf/nlregression.pdf NLS, a solver for non-linear regression problems .

    GAMS Modeling Notes

    http://www.gams.com/~erwin/nlp1.pdf Document with some NLP models.
    http://www.gams.com/~erwin/mip1.pdf Document with some MIP models.
    http://www.gams.com/~erwin/monalisa/monalisa.pdf Try to find the interesting parts of the Mona Lisa using an assignment problem.
    http://www.gams.com/~erwin/monalisa/monalisa.png The resulting picture.
    http://www.gams.com/~erwin/sos2/xy.pdf Global optimization of f(x,y) using SOS2 variables.
    http://www.gams.com/~erwin/book/gamsmodeling.pdf Work in progress: chapter 1 of text on Model building with GAMS. This chapter is on GAMS modeling.

    Linear programming

    http://www.gams.com/~erwin/book/lp.pdf Work in progress: chapter 2 of text on Model building with GAMS. This chapter is on linear programming.
    http://www.gams.com/~erwin/molp/molp.pdf Implementing Multi-objective/goal programming models in GAMS
    http://www.gams.com/~erwin/dea/dea.pdf Data Envelopment Analysis with GAMS. Some tricks with loops are shown that improve GAMS performance.
    http://www.gams.com/~erwin/lineq/lineq.pdf Solving systems of linear equations with GAMS. Some notes.
    http://www.gams.com/~erwin/dw/dw.pdf An annotated implementation of the Dantzig-Wolfe decomposition algorithm in GAMS

    Stochastic programming

    http://www.gams.com/~erwin/newsboy/newsboy.pdf The newsboy problem formulated as a two-stage stochastic linear program.
    http://www.gams.com/~erwin/twostage/twostage.pdf An annotated example of two-stage stochastic linear programming, with different implementations: deterministic equivalent, DECIS, OSL-SE and Benders Decomposition.
    http://www.gams.com/~erwin/stochbenders.pdf Solving a two-stage stochastic linear programming problem through Benders Decomposition.

    Integer programming

    http://www.gams.com/~erwin/book/mip.pdf Work in progress: chapter 3 of text on Model building with GAMS. This chapter is on integer programming.
    http://www.gams.com/~erwin/langford/langford.pdf Langford's Problem: a permutation problem solved as an integer programming model.
    http://www.gams.com/~erwin/benders/benders.pdf Fixed Charge Transportation Problem solved using Benders' Decomposition implemented in GAMS.
    http://www.gams.com/~erwin/lagrel/lagrange.pdf Example of Lagrangian Relaxation.
    http://www.gams.com/~erwin/colgen/colgen.pdf Column generation for the cutting stock problem.
    http://www.gams.com/~erwin/recr/tiling.pdf The tiling problem. Open question: Who can show that h(110)=1 using Mixed Integer Programming?
    http://www.gams.com/~erwin/recr/alphametics.pdf Another recreational problem: alphametics.
    http://www.gams.com/~erwin/ppp.pdf
    http://www.gams.com/~erwin/ppp_sanjose.pdf
    		 Experiments with an unsolvable MIP model. This is a longer version of paper to appear in Computers & Operations Research. Also slides from INFORMS 2002 San Jose talk.
    http://www.gams.com/~erwin/tsp/stsp42.pdf Symmetric traveling salesman problem coded in GAMS.
    http://www.gams.com/~erwin/tsp.gif Animated GIF illustrating the above 42 city TSP problem
    http://www.gams.com/~erwin/tsp/plot2.pdf Plots for the symmetric traveling salesman problem.

    Nonlinear programming

    http://www.gams.com/~erwin/inistat/inistat.pdf Description of INISTAT, a tool to get a quick overview of the bounds, scaling and initial values of an NLP model.
    http://www.gams.com/~erwin/inistat/inistat.zip Software tool itself. Requires GAMS 21.3 on PC/Windows.
    http://www.gams.com/~erwin/inistat/turkey.html Example output.
    http://amsterdamoptimization.com/pdf/nlregression.pdf Nonlinear regression solver. This is a GAMS solver that can solve nonlinear least squares problems and report a suite of standard regression statistics.

    Databases

    http://www.gams.com/~erwin/gamsqueue.pdf Description of GAMS/ORAQUEUE, a queueing package to submit GAMS jobs in an Oracle environment. A web based submission tool is used to submit GAMS jobs using a standard browser. The code can be used to develop robust Web based optimization applications.
    MDB2GMS A tool for importing MS Access tables into GAMS. For download instructions see http://www.gams.com/~erwin/interface/interface.html section WTOOLS.
    SQL2GMS A tool for importing data from SQL databases (Microsoft SQL Server, Oracle, IBM DB2 etc.) into GAMS. For download instructions see http://www.gams.com/~erwin/interface/interface.html section WTOOLS.
    GDX2ACCESS A tool for dumping a GDX file to an MS Access database. For download instructions see http://www.gams.com/~erwin/interface/interface.html section WTOOLS.
    http://www.gams.com/~erwin/mijdbc/mijdbc.html A JDBC driver for the first commercial object-relational database Illustra.
    http://amsterdamoptimization.com/packaging.html MS Office integration: call GAMS from MS Access.

    Micro-economics

    http://www.gams.com/~erwin/micro/budget.pdf Simple utility maximization with a twist. Nonlinear budgets formulated as a piecewise linear function. The resulting model can be solved with an MINLP solver.
    http://www.gams.com/~erwin/micro/oligopoly.pdf Oligopolistic producer behavior. A simple MCP model.
    http://www.gams.com/~erwin/micro/spatial.pdf Spatial partial equilibrium models in GAMS. Some simple examples.
    http://www.gams.com/~erwin/micro/ramsey.pdf A simple Ramsey growth model (this is a general equilibrium model).

    Statistics

    http://www.gams.com/~erwin/cenlim/cenlim.html A Java applet that demonstrates the Central Limit Theorem. This is a cornerstone result in Statistics.
    http://www.gams.com/~erwin/darts/darts.html Estimating pi by throwing darts.
    http://www.amsterdamoptimization.com/pdf/ols.pdf Many regression problems are in fact optimization problems (least squares, least absolute deviation, maximum likelihood, maximum entropy). We show here how GAMS can be used to solve these optimization problems.
    http://www.gams.com/~erwin/specfun.pdf New special functions GAMMA(.), BETA(.), LOGGAMMA(.), LOGBETA(.), GAMMAREG(.), BETAREG(.), BINOMIAL(.).
    http://amsterdamoptimization.com/pdf/regression.pdf Regression solver. This is a GAMS solver that can solve linear least squares problems and report a suite of standard regression statistics.
    http://amsterdamoptimization.com/pdf/nlregression.pdf Nonlinear regression solver. This is a GAMS solver that can solve non-linear least squares problems and report a suite of standard regression statistics.
    http://www.gams.com/~erwin/random.pdf Some notes on generating random numbers with GAMS.

    MINLP models

    http://www.gams.com/~erwin/gastrans/gastrans.pdf The MINLP model gastrans.gms in the model library can be solved very efficiently as a DNLP model.
    http://www.gams.com/~erwin/bb/bb.pdf Branch-and-bound methods for an MINLP model with semi-continuous variables. The GAMS code shows how to roll your own branch-and-bound algorithm.
    http://www.gams.com/~erwin/minlp/minlp.pdf Different MINLP algorithms: branch-and-bound (BB), Generalized Benders Decomposition (GBD) and Outer Approximation (OA) formulated in GAMS applied to a problem from finance: the cardinality constrained portfolio problem.