Generation of some first-order autoregressive Markovian sequences of positive random variables with given marginal distributions
Lawrance, A. J.
Lewis, P. A. W.
MetadataShow full item record
Methods for simulating dependent sequences of continuous positive-valued random variables with exponential uniform, Gamma, and mixed exponential marginal distributions are given. In most cases the sequences are first-order, linear autoregressive, Markovian processes. A very broad two-parameter family of this type, GNEAR(1), with exponential marginals and both positive and negative correlation is defined and its transformation to a similar multiplicative process with uniform marginals is given. It is shown that for a subclass of this two-parameter family extension to mixed exponential marginals is possible, giving a model of broad applicability for analyzing data and modelling stochastic systems, although negative correlation is more difficult to obtain than in the exponential case. Finally, two schemes for autoregressive sequences with Gamma distributed marginals are outlined. Efficient simulation of some of these schemes is discussed. (Author)
NPS Report NumberNPS55-81-003
Showing items related by title, author, creator and subject.
Lewis, P.A.W.; Lawrance, A.J. (1979);Methods of simulation depending sequences of continuous positive-valued random variables with exponential and uniform marginal distributions are given. In most cases the sequences are first-order, linear autoregressive, ...
Heavy traffic results for single server queues with dependent (EARMA) service and interarrival times Jacobs, Patricia A. (Monterey, California. Naval Postgraduate School, 1979-04); NPS55-79-009Models are given for sequences of correlated exponential interarrival and service times for a single server queue. These multivariate exponential models are formed as probabilitic linear combinations of sequences of ...
Lewis, Peter A.W. (1981);Recent work has made the generation of univariate time series for inputs to stochastic systems quite simple. The time series are all random linear combina- tion~ of i.i.d. random variables with Exponential, Gamma and ...