
207A Game of Competitive Investment
Our framework can be extended beyond the current bi-valued case. We
opt for the current case for the ease of later derivations involving controlled
information acquisition.
Now, we suppose that the information structure itself is subject to rms’
controls. That is, we allow the random variable Θ to be dependent on rms’
efforts. There are stages 0 and 1. In stage 0, rms may invest to learn the
market; in stage 1, they may participate in the investment game described in
Section 8.2. Before stage 0, all rms believe that the return function follows
R
Ω
(x). Given stage 0 rm-effort vector x
0
= (x
0i
|i = 1,2,…,