Abstract
We document, for a new data set, the existence of daily seasonality. The data set consists of the trades in four equities and two bonds on the Dublin stock exchange for the mid nineteenth century.
Acknowledgement
The authors wishes to thank the staff of the National Archives of Ireland and also Shane Whelan and Edel Tully for data and other advice.
Notes
1 These test procedures being well-known details of their construction are omitted, but are available from the authors on request.
2 We also estimated the impact of daily seasonality using least absolute deviation and least trimmed squares estimators, methods which are robust to outliers and have high breakdown points. The results were in all cases qualitatively the same.