Christian Conrad and Karin Loch, "Anticipating Long-Term Stock Market Volatility", Journal of Applied Econometrics, Vol. 30. No. 7, 2015, pp. 1090-1114. ***** DATA ***** The data are provided in four plain ASCII files in DOS format. The data files are zipped in cl-data.zip. The data are also provided in Excel spreadsheets and zipped in cl-data-xls.zip. Unix/Linux users should use "unzip -a" for cl-data.zip only. (1) cl-returndata.txt: daily log returns of the S&P 500 from Jan 2nd 1969 to Dec 30th 2011 (10852 observations) Daily return data were obtained from the Federal Reserve Bank of St. Louis database: http://research.stlouisfed.org/fred2/ (2) cl-macrodata.txt: quarterly realized volatilities and macro variables from 1969Q1 to 2011Q4 (172 observations) The macro variables included in this file are: * rvq: quarterly realized volatility (sum of daily squared returns) * drgdp: real GDP growth (annualized quarter-over-quarter percentage change) * dip: industrial production growth (annualized quarter-over-quarter percentage change) * dunemp: unemployment rate change * dhous: housing starts growth (annualized quarter-over-quarter percentage change) * dcprof: corporate profits growth (annualized quarter-over-quarter percentage change) * infl: GDP deflator inflation (annualized quarter-over-quarter percentage change) ** nai: National Activity Index ** no: ISM new orders ** dconss: Michigan Consumer Sentiment Index (change) ** drcons: real personal consumption growth (annualized quarter-over-quarter percentage change) ** tspread: Term spread (difference between the 10-year Treasury bond yield and the 3-month T-bill rate) Notes: * Variables that are included in the SPF dataset were obtained from the Federal Reserve Bank of Philadelphia database: http://www.philadelphiafed.org/research-and-data/real-time-center/survey-of-professional-forecasters/ For all variables except corporate profits, we consider the first release data. This corresponds to the variable "Realiz1" in the respective dataset for the SPF forecast error statistics. For drgdp, inflation, and drcons the 1995Q4 first release observation is not available and is replaced with the respective second release observation. In line with the SPF dataset, we consider nominal corporate profits excluding IVA and CCAdj before 2006 and including IVA and CCAdj since 2006Q1. ** all other variables were obtained from the Federal Reserve Bank of St. Louis database: http://research.stlouisfed.org/fred2/ (3) cl-spfdata.txt: quarterly SPF forecasts from 1969Q1 to 2011Q4 (172 observations) This file includes median forecasts for the quarter in which the survey is conducted (X_2) and forecasts up to four-quarters-ahead (X_6) for the following variables: drgdp, dip, dunemp, dhous, dcprof, and infl. The four-quarters-ahead forecasts X_6 are not available for the 1969Q1-Q3, 1970Q1, and 1974Q3 surveys. They are replaced by predictions based on X_2 to X_5, see also the data discussion in Section 3. (4) cl-rvintradata.txt: daily realized variance based on 5-min intradaily returns of the S&P 500 from Jan 3rd 2000 to Dec 31st 2010 (2767 observations) The file contains the daily realized variance measure which is used for the forecasting evaluation in Section 4.3. The data were obtained from the Oxford-Man Institute of Quantitative Finance, Realized Library: http://realized.oxford-man.ox.ac.uk/data ********* APPENDIX ********* The online Appendix contains additional tables and figures. In particular, all tables and figures related to the results in Section 4.4. are presented. The GARCH-MIDAS model estimations were implemented in OX. Programs are available upon request. Please adress any questions to: Karin Loch Heidelberg University karin.loch [AT] awi.uni-heidelberg.de