Parameter estimation for ARTA processes

Providing accurate and automated input-modeling support is one of the challenging problems in the application of computer simulation. The models incorporated in current input-modeling software packages often fall short of what is needed because they emphasize independent ana identically distributed processes, while dependent time-series processes occur naturally in the simulation of many real-life systems. This paper introduces a statistical methodology for fitting stochastic models to dependent time-series input processes. Specifically, an automated and statistically valid algorithm is presented to fit ARTA (autoregressive-to-anything) processes with marginal distributions from the Johnson translation system to stationary univariate time-series data. The use of this algorithm is illustrated via a real-life example.

[1]  Song Wheyming Tina,et al.  Generating pseudo-random time series with specified marginal distributions , 1996 .

[2]  D. Marquardt.  An Algorithm for Least-Squares Estimation of Nonlinear Parameters , 1963 .

[3]  J. J. Swain,et al.  Least-squares estimation of distribution functions in johnson's translation system , 1988 .

[4]  Mark E. Johnson.  Multivariate Statistical Simulation: Johnson/Multivariate , 1987 .

[5]  I. D. Hill,et al.  Fitting Johnson Curves by Moments , 1976 .

[6]  M. E. Johnson,et al.  Multivariate Statistical Simulation , 1988 .

[7]  Barry L. Nelson,et al.  A comprehensive input-modeling framework and software for stochastic, discrete-event simulation experiments , 2002 .

[8]  Miron Livny,et al.  The Impact of Autocorrelation on Queuing Systems , 1993 .

[9]  T. W. Anderson.  Goodness of Fit Tests for Spectral Distributions , 1993 .

[10]  Barry L. Nelson,et al.  Modeling and Generating Multivariate Time Series with Arbitrary Marginals Using a Vector Autoregress , 2000 .

[11]  Helen J. Mikulski,et al.  Statistical Computer Programs: A Sequential Simplex Program for Solving Minimization Problems , 1974 .

[12]  Barry L. Nelson,et al.  Automatic modeling of file system workloads using two-level arrival processes , 1998, TOMC.

[13]  P. A. Blight.  The Analysis of Time Series: An Introduction , 1991 .

[14]  N L JOHNSON,et al.  Bivariate distributions based on simple translation systems. , 1949, Biometrika.

[15]  T. Lai.  Time series analysis univariate and multivariate methods , 1991 .

[16]  Jacek Blazewicz,et al.  The job shop scheduling problem: Conventional and new solution techniques , 1996 .

[17]  Maurice G. Kendall,et al.  The advanced theory of statistics , 1945 .

[18]  George S. Fishman,et al.  Solution of Large Networks by Matrix Methods , 1976, IEEE Transactions on Systems, Man, and Cybernetics.

[19]  James R. Wilson,et al.  Least squares estimation of nonhomogeneous poisson processes , 2000 .

[20]  William W. S. Wei,et al.  Time series analysis - univariate and multivariate methods , 1989 .

[21]  Huifen Chen,et al.  Initialization for NORTA: Generation of Random Vectors with Specified Marginals and Correlations , 2001, INFORMS J. Comput..

[22]  N. L. Johnson,et al.  Systems of frequency curves generated by methods of translation. , 1949, Biometrika.

[23]  Barry L. Nelson,et al.  Numerical Methods for Fitting and Simulating Autoregressive-to-Anything Processes , 1998, INFORMS J. Comput..

[24]  C. L. Mallows,et al.  Linear processes are nearly Gaussian , 1967 .

[25]  Barry L. Nelson,et al.  Autoregressive to anything: Time-series input processes for simulation , 1996, Oper. Res. Lett..

[26]  David Goldsman,et al.  The TES methodology: modeling empirical stationary time series , 1992, WSC '92.