This file explains how to generate the results in "Using Long-run Consumption-Return Correlation to Test Asset Pricing Models",
by Jianfeng Yu.


1, Use file Habit_IID_Wachter.m for habit model with IID Growth. 
It helps generate results in Talbes 1 and 4 and the results in Figures 3 and 7.
Please see the comments in the file for details.


2, Use file Bansal_Yaron_model.m for full Bansal and Yaron (2004) model with stochastic volatility. 
It helps generate Figures 6 and 7. Please see the comments in the file for details.


3, Use file B_Y_no_Vol.m  for the analytical example on the long-run risk model without stochastic volatility.
It helps generate Figures 4, 5, and 7. Please see the comments in the file for details.


4, Use file coh_data_2011_neat.m  for the analysis on historical data.


Details: This program produce the Granger-Causality Test, Long-run correlation,
Bandpass filter analysis, and Spectrum analysis for consumption and excess 
return data. It produces figure 1 and figure 2 in the paper 
and it produces the results in Section 2 of the paper.  Some of the input files are coming out of simulation of the models.

Note: There are also a few auxiliary files used in our calculation including BPF.m, ols.m etc. 
Thus to run the program, one need to have those files as well.  (Those files are provided.)

5, Finally, the stata.do file is used to generate the estimation in Table 3.