LHS vs RHS sampling advantages, sum-to-zero QR based

andre.pfeuffer · July 31, 2019, 1:00pm

Hi Ben,

sure the link is here and following posts:

Meanwhile I’ve tested both and the results look they do the same. The LHS could be used for an internal Stan sampling statement.

I always asked myself, why should somebody use a transformation on the LHS, when it can be done on the RHS? Is there some efficiency bonus?

Topic		Replies	Views
Sum-to-zero and sbc Modeling	0	485	March 1, 2022
QR decomposition and reduce_sum undercutting each others' runtime gains? Modeling techniques , performance , cmdstanr	6	787	June 2, 2020
State-space model (vectorized sampling vs for loop with recursion?) Modeling	3	591	September 25, 2018
Help with the explanation of drawing sample from the loop - Vector verses Real Modeling techniques , specification	2	369	December 15, 2020
What's the difference between putting all the model formula into transformed parameters block and leaving the model formula into the model block? Modeling specification , performance	4	526	September 20, 2021