Table 4.

Proposal distributions.

Param.	Proposal Distribution	Tuning Param.	MH Ratio
κ	κ* = κ exp [κt(U - 0.5)], U ~ Unif(0, 1)	κt = 1.0
α	α* ~ Normal(α, αt)	αt = 0.5
t i	= ti exp [tt(U - 0.5)], U ~ Unif(0, 1)	tt = 1.0
θ	θ* ~ Normal(θ, θt)	θt = 0.2
	l* ~ Normal(l, lt)	lt = 0.2
	b* ~ Unif(1,...,2N - 3), l* ~ Unif(0, 1)	NA
d	d* = 1 - d	NA

For each parameter listed, a new value (starred, e.g. κ*) is proposed according to the listed distribution. Branch lengths t_i are updated one at a time for i = 1,...,2N - 3. Tuning parameters are subscripted by t. Updates of either l or (b, l) are mixed with probability m_t. Update of d is a trans-dimensional move. The Metropolis-Hastings acceptance ratios are given in the last column. Dependence on parameters not involved in the update is not shown.