When s 2 confirms s 2 (either s 1 =H, s 2 =Uor s 1 =L, s 2 =D), the player be-
comes overconfident and acts as if his precision were pCinstead of p, so
(A15)When s 2 is informative (q > 1/2), this probability exceeds pC. When s 2 does
not confirm s 1 , the player does not become overconfident, so
(A16)When evaluated with an informative signal s 2 (q>1/2), this probability is
less than p. With a risk neutral player, the price of the asset with value θ
can be calculated linearly using the above probabilities. The price at time
0 (P 0 ) is, by definition, equal to 0. As θcan take on a value of +1 or −1,
the price is (ρ)(+1)+(1−ρ)(−1) or, 2ρ−1, where ρis the probability that
θis +1.
(A17)(A18)(A19)The price changes are ∆P 1 =P 1 −P 0 =P 1 and ∆P 2 =P 2 −P 1. E[P 1 ]=0, so
cov(∆P 1 , ∆P 2 )=E[∆P 1 ∆P 2 ]. The probabilities of the eight possible outcomes
are:
Pr(θ=+1, s 1 =H, s 2 =U)=Pr(θ=−1, s 1 =L, s 2 =D)=pq/2 (A20)
Pr(θ=−1, s 1 =H, s 2 =U)=Pr(θ=+1, s 1 =L, s 2 =D)
=(1−p)(1−q)/2 (A21)Pr(θ=+1, s 1 =H, s 2 =D)=Pr(θ=−1, s 1 =L, s 2 =U)=p(1−q)/2 (A22)
Pr(θ=−1, s 1 =H, s 2 =D)=Pr(θ=+1, s 1 =L, s 2 =U)=(1−p)q/2. (A23)
PP sHsD
pq
pq pq22 Hs D 22 Hs U^211212ss 11 ,,Pr(,).===− === =+ = = −=−
+−θPP sHsU
pq
pq qHs U Hs D
C
C222221121
1
211ss 11 ,,Pr(,)()()===− === =+ = = −=+−
−+−θPP 11 ss 11 ==HL=− =21 121Pr()θ=+sHp 1 = − = −PrPr Pr
Pr(,)(, )()
(, )
()
()().θθθ
=+ = = ====+=+
===−
−+ −1111
111212
12
sHsDsHsD
sHsD
pq
pqqpPrPr Pr
Pr(,)(, )()
(,)()().θθθ
=+ = = ====+=+
===
−+−11121112
12
12
sHsUsHsU
sHsU
pq
pq qC
CINVESTOR PSYCHOLOGY 493