Luiz Lima and Fanning Meng, "Out-of-Sample Return Predictability: A 
Quantile Regression Approach", Journal of Applied Econometrics, Vol. 32, 
No. 4, 2017, pp. 877-895.

All data and programs are zipped in the file lm-files.zip. All files
are ASCII files in DOS format. Unix/Linux users should use "unzip -a".

The data come from Amit Goyal's webpage
(http://www.hec.unil.ch/agoyal/), including monthly returns on the S&P
500 index, the risk-free rate, and 15 predictors from December 1926 to
December 2013.  In total, there are 1045 observations. A detailed
description of each predictor is available in the "data description"
section of the online appendix file.

The data are stored in a .csv file named "Goyal_Welch_RF.csv", size of
196kB. The first column is the date variable "yyyymm", where "yyyy"
and "mm" stands for year and month respectively.  Column 2 is a vector
of ones, representing the constant term. Columns 3-17 are predictors.
Column 18 is returns on S&P 500 index. Column 19 is risk-free returns.

The main codes are stored in the file named "Lima_Meng_codes.R". The
other code file for figure 5 is named "Fig5_codes_paper.m". Codes for
the appendix are in the file Lima-Meng-Appendix_codes.R.

Luiz Lima            and   Fanning Meng
llima [AT] utk.edu         fmeng1 [AT] vols.utk.edu