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