Jaron T. Krogel

I request we add stabilization of the batched DMC algorithm before it becomes default. See also #3082 and #3608.

Other issues: #3522

I like this as a per-optimizer block feature. The only change/request I would make is to not innovate on adding lots of new xml tags. Almost all other inputs to...

@ye-luo it does for 1RDM: ``` spo_u spo_uv matrix uniform_grid 4 1.0 0 0 0 ``` I suggest we update parameter handling (if needed) to handle lists of the basic...

Propose a weighted autocorrelation algorithm and we're in business.

I would like to move to unbiased weighted statistics in a consistent way throughout qmca

Not really. The practical use cases (VMC, production DMC) have no or very little weight fluctuation and very little bias (like 1/100 of the statistical errorbar) when analyzed in practice....

Yes, we definitely need to account for weights. We just don't have the formulae to account for them yet, therefore this is the first task.

Formulae for weighted mean and equal time covariance: https://en.wikipedia.org/wiki/Weighted_arithmetic_mean#Weighted_sample_covariance See "Weighted sample covariance" and "Frequency weights".

What is missing is autocorrelation time estimation with weighted time series. Probably a biased formula for this can be derived in the discrete (weighted delta function) limit from formulae for...