Michael P. Clements and Nick Taylor, "Evaluating Interval Forecasts of High-Frequency Financial Data", Journal of Applied Econometrics, Vol. 18, No. 4, 2003, pp. 445-456. All data and program files are ASCII files in DOS format. They are zipped in the file ctfiles.zip. Unix users should use "unzip -a". The data are contained in two files, hstock.data and hcondv.data. The file hstock.data consists of a 1673 by 10 matrix with the columns representing the different variables and the rows representing the time series observations. The observations are hourly in frequency and cover the period 2/1/98 to 29/12/98. The first variable in the (log) return on the (nearest) FTSE100 futures contract and the second variable is the volume of trading in this contract. Both of these series were obtained from LIFFE. The third variable indicates the hour of trade, and the remaining seven variables are hourly dummy variables. hcondv.data consists of a 1673 by 8 matrix that is arranged in a similar fashion to hstock.data. The first two columns contain the standardized residuals and conditional volatility obtained from estimating a GARCH(1,1) model. The remaining columns are similarly arranged and contain the standardized residuals and conditional volalities from a GARCH(1,1)-V model, a PGARCH(1,1) model, and a PGARCH(1,1)-V model, respectively. A sample GAUSS program, spec0.prg, is also supplied. It calculates the rejection probabilities associated with the various tests considered in the paper.