changes τperiods in the future ∆Pt+τ=Pt+τ−Pt+τ− 1. These correlations
are then averaged over the Monte Carlo draws. The average correlations
are plotted in figure 13.4. This simulation yields:
Result 4.In the biased self-attribution setting of subsection B, short-
lag correlations between single-period stock price changes and past
earnings are positive, and long-lag correlations can be positive or
negative.To summarize, the analysis suggests that the conclusion from the basic
model that investors overreact to private signals holds in the dynamic
model. While investors underreact on average to public signals, public sig-
nals initially tend to stimulate additional overreaction to a previous pri-
vate signal. Thus, underreaction is mixed with continuing overreaction.
In the model of this section, earnings-based return predictability and mo-
mentum both arise from self-attribution bias. Further, the literature cited in
subsection B.3 suggests that the magnitude of this bias varies systematically
across countries. Based on these observations, the self-attribution model
suggests a positive relationship across international markets between the
strength of the momentum effect and that of the postearnings announce-
ment drift.
486 DANIEL, HIRSHLEIFER, SUBRAHMANYAM
Figure 13.4. Correlation between information changes and future price changes.
This figure shows the set of average sample correlations between the ∆etand price
changes τperiods in the future ∆Pt+τ=Pt+τ−Pt+τ− 1. These are calculated using
the simulated dynamic model of section 3.B.3.