Claudia Foroni and Massimiliano Marcellino, "Mixed Frequency Structural Models: Identification, Estimation, and Policy Analysis", Journal of Applied Econometrics, Vol. 29, No. 7, 2014, pp. 1118-1144. The data covers the period January 1965 - December 2007. The data have been downloaded from FRED, St. Louis FED. List of series: - Real Gross Domestic Product, 3 Decimal (GDPC96) - Consumer Price Index for All Urban Consumers: All Items (CPIAUCSL) - Personal Consumption Expenditures (PCEC) - Fixed Private Investment (FPI) - Civilian Employment (CE16OV) - Effective Federal Funds Rate (FEDFUNDS) - Civilian Noninstitutional Population (CNP16OV) - Nonfarm Business Sector: Average Weekly Hours (PRS85006023) - Average Weekly Hours Of Production And Nonsupervisory Employees: Total private (AWHNONAG) For the estimation of the models, we use the data as in Smets and Wouters, 2007 (following their transformations). The quarterly variables used are: - log difference of real GDP - log difference of real consumption - log difference of real investment - log difference of real wage - log hours worked - the log difference of the CPI index - federal fund rate. In the mixed-frequency model, we use monthly values of FFR, the log difference of the consumer price index and hours. The log difference of real GDP, real con- sumption, real investment and real wage stay the same. ***** 1. DATA The file fm-data.zip contains files with the data used in the paper. The data are stored as ASCII files in DOS format. Unix/Linux users should use "unzip -a". The variables are stored as columns. - dataQ: Contains the quarterly data as downloaded from FRED - dataQtr: Contains the quarterly data as used in the estimation (transformed accordingly to the transformations in Smets and Wouters 2007) - dataM: Contains the monthly data as downloaded from FRED - dataMFtr: Contains the mixed-frequency (monthly and quarterly) data as used in the estimation (transformed accordingly to the transformations in Smets and Wouters 2007) ***** 2. APPENDIX The Online Appendix contains the equations of the model and the additional graphs.