Conditional on βn, the probability that a manager of a facility with Tn regulated
units makes the observed Tn compliance choices is:
τn
PUn = i) ≡ pRpt(β) = ∏
t=l
eβn'xnU
(7)
Jnt
∑eβJχnjt
J=l
where i is a Tn * 1 dimensional vector denoting the sequence of observed choices.
Unconditional choice probabilities are derived by integrating conditional choice prob-
abilities over the distribution of unobserved random parameters (Train, 2003). The
θ vector of unknown parameters describes the distribution of β. The parameter esti-
mates are those that maximize the following log likelihood function:
ɪ 7 jn eβ'χn<t
LL(θ) = ∑ln ∏j ------f (β ∖θ)dβ,
(8)
n=l t t=l я'^
—ɔo ∖ eβ χnjt
J=l
Jnt is the number of viable compliance alternatives available to unit t operated by
manager n. Because this integral does not have a closed form solution, the uncon-
ditional probabilities are approximated numerically through simulation. For each
decision maker, R draws of β are taken from the density f (β ∣θ)jone for each decision
maker. For each draw, the value of [7] is calculated for each decision maker. The
results are averaged across draws. Simulated maximum likelihood estimates of the
parameters maximize the following:.
N R Tn
SLL(θ') = £ ln I £||
n=l r=l t=l
eβn'≈n>t
Jnt
(9)
∑ef,t⅛t
J=l
To increase the accuracy of the simulation, 1000 pseudo-random Halton draws are
20