Worapree Maneesoonthorn, Catherine S. Forbes, and Gael M. Martin,
"Inference on Self-Exciting Jumps in Prices and Volatility using
High-Frequency Measures", Journal of Applied Econometrics, Vol. 32,
No. 3, 2017, pp. 504-532.

All data are in the file mfm-data.txt, an ASCII (CSV) file in DOS
format. It is zipped in the file mfm-data.zip. Unix/Linux users should
use "unzip -a".

The dataset contains the daily financial return, realized variance,
bipower variation, and various price jump measures for the S&P500
market index between January 3rd 1996 and June 23rd 2014. The high
frequency index data were supplied by the Securities Industries
Research Centre of Asia Pacific (SIRCA) on behalf of Reuters, with the
raw intraday index data cleaned using methods similar to those of
Brownlees and Gallo (2006). 

The measures constructed from high-frequency data are based on fixed
five minute sampling, with a "nearest price" method (Andersen et al.,
2007) used to construct the relevant returns five minutes apart, and
only index values recorded within the New York Stock Exchange market
trading hours used in the construction.

The columns of data are described below:

Column 1:  "Date"   - records the date in dd-mm-yy format.

Column 2:  "Return" - daily return, annualized using a scale factor of 250.

Column 3:  "RV"     - realized variance computed as per equation (16) in
                      the paper, annualized using a scale factor of 250.

Column 4:  "BV"     - bipower variation computed as per equation (17) in
                      the paper, annualized using a scale factor of 250.

Column 5:  "Ip"     - price jump occurrence measure computed as per equation
                      (18) in the paper.

Column 6:  "TQ"     - tripower quarticity used to construct the price jump
                      test statistic (equation (19)), annualized using a
		      scale factor of 250.

Column 7:  "Mptil"  - log price jump magnitude measure, computed as per
                      equation (20) in the paper. Here, N/A corresponds to
		      cases where Zptil=0. This quantity is expressed as the
		      log of annualized price jump magnitude to match the
		      scale of the annualized returns in Column 2. Refer to
		      the description of the data in Column 8 for more details.

Column 8:  "Zptil"  - price jump magnitude measure, computed as per equation
                      (21) in the paper, scaled by sqrt(250) to match the
		      scale of the annualized returns in Column 2.


References:


1. Brownlees, C.T. and Gallo, G.M. 2006. "Financial Econometric
Analysis at Ultra-High Frequency: Data Handling Concerns",
Computational Statistics and Data Analysis, 51, 2232-2245.
 
2. Andersen, T.G., Bollerslev, T. and Diebold, F.X. 2007. "Roughing It
Up: Including Jump Components in the Measurement, Modeling and
Forecasting of Return Volatility", The Review of Economics and
Statistics, 89, 701-720.