model{
        # Likelihood

        for(i in 1:n.rec){

        z[i] <- equals(trips.star[i], 0)        # I(visits = 0)
        logit(q[i]) <- gamma[1]*intercept.rec[i] + gamma[2]*ski[i] +
                gamma[3]*fee[i] + gamma[4]*income[i]
        pi[i] <- max(0.001, min(0.999, q[i]))
        mu[i] <- beta[1]*intercept.rec[i] +b eta[2]*ski[i] + beta[3]*fee[i] +
                beta[4]*income[i] + (1 - 2*tau)*w[i]/(tau*(1 - tau))
        prec[i] <- delta*tau*(1 - tau)/(2*w[i])
        log.like[i] <- z[i]*log(1 - pi[i]) + (1 - z[i])*(log(pi[i]) +
                0.5*log(prec[i]/(2*3.14159)) - 0.5*pow(trips.star[i] - mu[i], 2)*prec[i])

        # zeros trick

        zeros[i] <- 0
        zeros[i] ~ dpois(lambda[i])
        lambda[i] <- -log.like[i] + 10000
        }

        # prior distributions

        for(i in 1:4){

        beta[i] ~ dnorm(0, 0.001)

        gamma[i] ~ dnorm(0, 0.001)
	}
        delta ~ dgamma(0.1, 0.1)

        for(i in 1:n.rec){w[i] ~ dexp(delta)}
}