The estimation of the Barndorff-Nielsen and Shephard model from daily data based on measures of trading intensity

Artikel i vetenskaplig tidskrift, 2008

We give a method to fit the Barndorff-Nielsen and Shephard model [J. R. Stat. Soe. Ser. B 2001; 63:167-241] to daily data. Many researchers have established a connection between volatility and different measures of trading intensity, such as traded volume or number of trades. We benefit from this connection, and propose to use some measure of trading intensity as the volatility in the model in [J. R. Stat. Soe. Ser. B 2001; 63:167-241] . Our approach gives stable parameter estimates, and it is much easier to implement than the quadratic variation method. The efficiency of our method is illustrated by a statistical analysis on the Ericsson stock from the OMX Stockholmsbörsen during an exceptionally turbulent period of five years. The results indicate a good model fit. Copyright © 2007 John Wiley & Sons, Ltd.