GARCH Time Series Models: An Application to Retail Livestock Prices

Satheesh V. Aradhyula, Matthew T. Holt
May 1988  [88-WP 29]

Traditional time series models assume a constant conditional variance. Realizing the implausibility of this assumption, Bollerslev proposed Generalized Autoregressive Conditional Heteroscedasticity (GARSH) processes, which are characterized by nonconstant conditional variances. In this paper, GARCH (1,1) processes were applied to model livestock prices. Results indicate that GARCH processes adequately describe retail meat price behavior.

Full Text 457 kB