Article

Article title MODIFICATION OF THE METHOD OF MOMENTS FOR ESTIMATION PARAMETERS OF THE BOX–JENKINS MODELS
Authors V.S. Vasiliev
Section SECTION VI. MATHEMATICAL MODELING IN THE STUDY OF BIOLOGICAL OBJECTS
Month, Year 04, 2013 @en
Index UDC 330.43
DOI
Abstract The autoregressive and moving average models by Box and Jenkins become since time of its appearance a standard instrument of analysis of time series. In the method of moments autoregressive component parameters are valued from extended Yule-Wolker equations. The decision of this linear system of equations in practice problems does not cause. The decision of nonlinear system equations for estimation moving average component parameters by method Newton, opposite, presents the problem, on which indicated authors themselves. In article is offered and researched iterative procedure for decision of the nonlinear system of the method of moments. Condition to convergence iterative process are received, showing influence values parameter themselves on velocity of convergence.

Download PDF

Keywords Econometrics; time series models; forecasting; Yule–Walker equations; Box–Jenkins models; method of moments; convergence iterative processes.
References 1. Box G.E.P. and Jenkins G.M. Time Series Analysis: Forecasting and Control, rev. Ed., San Francisco: Holden-Day, 1976.

Comments are closed.