Overspecified Asset Pricing Models JFE
Description
YO_DATA0.xlsx - Data for Tables 1 and 2. Quarterly data 1951-2001. Sheet 1 "r25" - 25 nominal excess returns of size and book-to-market sorted portfolios. Sheet 2 "YO" - 3 factors in Yogo model (market, nondurables, durables). The above data were taken from Motohiro Yogo's web page https://sites.google.com/site/motohiroyogo/ YO_DATA.xlsx - Data for Tables 1 and 2. Quarterly data 1949-2012. Sheet 1 "r5" - 5 real excess returns of industry portfolios. Sheet 2 "r6" - 6 real excess returns of size and book-to-market sorted portfolios. Sheet 3 "r25" - 25 real excess returns of size and book-to-market sorted portfolios. Sheet 4 "YO" - 3 factors in Yogo model (market, nondurables, durables). The above data were kindly provided by Craig Burnside. JW_DATA.xlsx - Data for Tables 1 and 2. Monthly data 1959-2012. Sheet 1 "r5" - 5 nominal excess returns of industry portfolios. Sheet 2 "r6" - 6 nominal excess returns of size and book-to-market sorted portfolios. Sheet 3 "r25" - 25 nominal excess returns of size and book-to-market sorted portfolios. Sheet 4 "JW" - 3 factors in Jagannathan-Wang model (market, labor, premium). The above data were kindly provided by Cesare Robotti. In all cases, the original return data were taken from Kenneth French data library https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html
Files
Steps to reproduce
FOLDER TABLES135: Replication of Tables 1, 3, and 5. These tables report results for CU-GMM. Files: - j1fk_cu.prg - Gauss code for j test d=1 with k factors (a general number of factors). - j2f3_cu.prg - Gauss code for j test d=2 with 3 factors. - j3f3_cu.prg - Gauss code for j test d=3 with 3 factors. - Rest of files - Output from previous codes, where "outXY" means output for Table X Panel Y. The researcher can choose the Table and Panel to replicate with the variable "tapa" at the beginning of each code. @tapa=1 for table 1@ @tapa=31 for table 3, panel a@ @tapa=32 for table 3, panel b@ @tapa=51 for table 5, panel a@ @tapa=52 for table 5, panel b@ The code "j2f3_cu" only provides the test of zero deltas for the common factor (the first factor) in the two basis SDFs. In the current ordering, this means the market factor for all the models. If researchers want to compute the zero deltas test for one of the other two factors, they just need to change the order of the factors, setting the corresponding factor as the first one. ------------------------------------------------- FOLDER TABLES246: Replication of Tables 2, 4, and 6. These tables report results for 2S-GMM and IT-GMM. Files: - jdf3_it.prg - Gauss code for j test d=1, 2, or 3 with 3 factors. - Rest of files - Output from previous code, where "outXY" means output for Table X Panel Y. The researcher can choose the Table and Panel to replicate with the variable "tapa" at the beginning of each code. @tapa=1 for table 1@ @tapa=31 for table 3, panel a@ @tapa=32 for table 3, panel b@ @tapa=51 for table 5, panel a@ @tapa=52 for table 5, panel b@ The particular d to test is chosen with the variable "d_dim".