Time Series Simulation with Quasi Monte Carlo Methods

URL:	http://www.math.psu.edu/ccma/Reports/Publications/...
Additional URL:	http://fmwww.bc.edu/cef00/papers/paper151.pdf
Document Type:	Working paper
Year of publication:	2000
The title of a journal, publication series:	Report / Penn State University, Department of Mathematics
Volume:	AM224
Place of publication:	Mannheim [u.a.]
Publication language:	English
Institution:	School of Law and Economics > VWL (Franz 1997-2013, Em)
Subject:	330 Economics
Subject headings (SWD):	Monte-Carlo-Simulation
Abstract:	This paper compares quasi Monte Carlo methods, in particular so-called (t,m,s)-net, with classical Monte Carlo approaches for simulating econometric time-series models. Quasi Monte Carlo methods have found successful application in many fields, such as physics, image processing, and the evaluation of finance derivatives. However, they are rarely used in econometrics. Here, we apply both traditional and quasi Monte Carlo simulation methods to time-series models that typically arise in macroeconometrics. The numerical experiments demonstrate that quasi Monte Carlo methods outperform traditional ones for all models we investigate.
Additional information:	Auch als: Computing in Economics and Finance 2000 ; 151