R-hat

Calculates an R-hat (potential scale reduction factor) value.

# S3 method for class 'mcmcr'
rhat(x, by = "all", as_df = FALSE, na_rm = FALSE, ...)

Arguments

x

An object.

by

A string indicating whether to determine by "term", "parameter" or "all".

as_df

A flag indicating whether to return the results as a data frame versus a named list.

na_rm

A flag specifying whether to ignore missing values.

...

Other arguments passed to methods.

Value

A number >= 1 indicating the rhat value.

Details

By default the uncorrected, unfolded, univariate, split R-hat value.

References

Gelman, A., and Rubin, D.B. 1992. Inference from Iterative Simulation Using Multiple Sequences. Statistical Science 7(4): 457–472.

See also

Other convergence: converged_pars(), converged_terms(), converged(), esr_pars(), esr_terms(), esr(), rhat_pars(), rhat_terms()

Examples

rhat(mcmcr_example)
#> [1] 2.002
rhat(mcmcr_example, by = "parameter")
#> $alpha
#> [1] 2.002
#> 
#> $beta
#> [1] 1.147
#> 
#> $sigma
#> [1] 1
#> 
rhat(mcmcr_example, by = "term")
#> $alpha
#> [1] 2.002 2.002
#> 
#> $beta
#>       [,1]  [,2]
#> [1,] 1.147 1.147
#> [2,] 1.147 1.147
#> 
#> $sigma
#> [1] 1
#> 
rhat(mcmcr_example, by = "term", as_df = TRUE)
#> # A tibble: 7 × 2
#>   term       rhat
#>   <term>    <dbl>
#> 1 alpha[1]   2.00
#> 2 alpha[2]   2.00
#> 3 beta[1,1]  1.15
#> 4 beta[2,1]  1.15
#> 5 beta[1,2]  1.15
#> 6 beta[2,2]  1.15
#> 7 sigma      1