Replication of tables and figures from "Belief Overreaction and Stock Market Puzzles" in Stata, R and Python.
Abstract:
We construct an index of long term expected earnings growth for S&P500 firms and show that it has remarkable power to jointly predict future errors in these expectations and stock returns, in both the aggregate market and the cross section. The evidence supports a mechanism whereby good news cause investors to become too optimistic about long term earnings growth, for the market as a whole but especially for a subset of firms. This leads to inflated stock prices and, as beliefs are systematically disappointed, to subsequent low returns in the aggregate market and for the subset of firms. Overreaction of long-term expectations helps resolve major asset pricing puzzles without time series or cross-sectional variation in required returns.
Paper Link: https://scholar.harvard.edu/shleifer/publications/expectations-fundamentals-and-stock-market-puzzles
A bold attempt to replicate the tables and figures of the paper "Belief Overreaction and Stock Market Puzzles" (BGLS) in the following languages:
- Stata
- R
- Python
Reference:
Bordalo P, Gennaioli N, La Porta R, et al. Belief overreaction and stock market puzzles[M]. National Bureau of Economic Research, 2020.