Abstract

<p>In this paper we introduce a latent variable based model for the dynamics of financial range, the stochastic <br>conditional range (SCR). We propose to estimate its parameters by Kalman filter, indirect inference and simulated <br>maximum likelihood depending on the hypotheses on the distributional form of the innovations. The model is <br>applied to a large subset of the S&amp;P 500 components. A comparison of its fitting and forecasting abilities with the <br>conditional autoregressive range (CARR) model shows that the new approach can provide a competitive alternative</p>