Ordinal-Response GARCH Models for Transaction Data: A Forecasting Exercise

Abstract

Using high-frequency transaction data, we evaluate the forecasting performance of several dynamic ordinal-response time series models with generalized autoregressive conditional heteroscedasticity. The specifications account for three components; leverage effects, in-mean effects and moving average error terms. To estimate the model parameters we develop Markov chain Monte Carlo algorithms. Our empirical analysis showed that ordinal-response models achieve better point and density forecasts than standard benchmarks, when they incorporate at least one of the three components.