We consider a general model in 1114. When we consider ns-1 breaks and dim states in
each, we have the following dimensions of matrices:
nps: ns x 1
tms: (ns x dim) x dim
paras: (ns x dim) x (# of parameters changing in the model).
When we set a scalar to nps and one to tms together with a single row of paras, we can
replicate the program leading to a steady state.
When we set same sets of parameters to paras by ns sets, we can replicate the program
with the same sets of parameters in 1115.
When we set the same tm’s to tms by ns, we can replicate the program with a single
transition matrix as in 1116.