pda.neff.calc.RdAutocorelation correction and efficient sample size calculation on latent process
pda.neff.calc(inputs, recalculate = FALSE)
| inputs | list |
|---|
inputs list, updated inputs with n_eff
What we're trying to do is to calculate the autocorrelation of the latent state, after attempting to "remove" the observation error. The first step is thus to estimate the latent state using a simple 'process free' state-space model (e.g. random walk).