Skip to contents

Estimate Evidence using a Nested Sampling Run

Source: R/calculate.R
calculate.ernest_run.Rd

Computes evidence and related quantities from a nested sampling run, optionally by simulating the volumes of each nested likelihood shell.

Usage

# S3 method for class 'ernest_run'
calculate(x, ..., ndraws = NULL)

Arguments

x

An ernest_run object.

...

These dots are for future extensions and must be empty.

ndraws

An optional positive integer. The number of log-volume sequences to simulate. If equal to zero, no simulations will be made, and a one draw vector of log-volumes are produced from the estimates contained in x. If NULL, getOption("posterior.rvar_ndraws") is used (default 4000).

Value

A tibble::tibble(), containing n_iter + n_points rows and several columns:

  • log_lik: The log-likelihood of the model.

  • log_volume: The log-volume of the prior space.

  • log_weight: The log weights of the live points.

  • log_evidence: The log-evidence of the model.

  • log_evidence_err: The standard error of the log-evidence (only available when ndraws = 0).

The tibble has the additional class ernest_estimate, which has its own plot method.

Each column is returned as an posterior::rvar() vector.

References

Higson, E., Handley, W., Hobson, M., & Lasenby, A. (2019). Nestcheck: Diagnostic Tests for Nested Sampling Calculations. Monthly Notices of the Royal Astronomical Society, 483(2), 2044–2056. doi:10.1093/mnras/sty3090

Examples

# Load an example run
data(example_run)

# View results as a tibble with `ndraws = FALSE` (the default).
calculate(example_run)
#> evidence estimates <ernest_estimate>
#> 
#> No. of Simulated Draws: 4000
#> Log. Volume: -17 ± 1.3
#> Log. Evidence: -9.1 ± 0.069
#> # A tibble: 10,398 × 4
#>       log_lik         log_volume   log_weight log_evidence
#>    <rvar[1d]>         <rvar[1d]>   <rvar[1d]>   <rvar[1d]>
#>  1  -147 ± NA  -0.00099 ± 0.0010  -154 ± 0.81  -154 ± 0.81
#>  2  -141 ± NA  -0.00199 ± 0.0014  -148 ± 0.80  -148 ± 0.80
#>  3  -136 ± NA  -0.00301 ± 0.0017  -143 ± 0.81  -143 ± 0.80
#>  4  -136 ± NA  -0.00402 ± 0.0020  -143 ± 0.82  -142 ± 0.67
#>  5  -134 ± NA  -0.00504 ± 0.0022  -141 ± 0.79  -140 ± 0.66
#>  6  -130 ± NA  -0.00604 ± 0.0025  -137 ± 0.79  -137 ± 0.75
#>  7  -130 ± NA  -0.00701 ± 0.0027  -137 ± 0.81  -136 ± 0.63
#>  8  -129 ± NA  -0.00804 ± 0.0028  -136 ± 0.81  -135 ± 0.58
#>  9  -129 ± NA  -0.00902 ± 0.0030  -136 ± 0.81  -135 ± 0.53
#> 10  -126 ± NA  -0.01000 ± 0.0031  -133 ± 0.80  -133 ± 0.69
#> # ℹ 10,388 more rows

# Generate 100 simulated log-volume values for each iteration.
calculate(example_run, ndraws = 100)
#> evidence estimates <ernest_estimate>
#> 
#> No. of Simulated Draws: 100
#> Log. Volume: -17 ± 1.2
#> Log. Evidence: -9.1 ± 0.076
#> # A tibble: 10,398 × 4
#>       log_lik        log_volume   log_weight log_evidence
#>    <rvar[1d]>        <rvar[1d]>   <rvar[1d]>   <rvar[1d]>
#>  1  -147 ± NA  -0.0013 ± 0.0013  -154 ± 0.92  -154 ± 0.92
#>  2  -141 ± NA  -0.0022 ± 0.0017  -148 ± 0.77  -148 ± 0.77
#>  3  -136 ± NA  -0.0032 ± 0.0019  -143 ± 0.65  -143 ± 0.64
#>  4  -136 ± NA  -0.0041 ± 0.0020  -143 ± 0.95  -142 ± 0.62
#>  5  -134 ± NA  -0.0050 ± 0.0023  -141 ± 0.89  -141 ± 0.72
#>  6  -130 ± NA  -0.0060 ± 0.0027  -137 ± 0.80  -137 ± 0.78
#>  7  -130 ± NA  -0.0070 ± 0.0030  -137 ± 0.79  -136 ± 0.67
#>  8  -129 ± NA  -0.0081 ± 0.0031  -136 ± 0.76  -135 ± 0.60
#>  9  -129 ± NA  -0.0092 ± 0.0031  -136 ± 0.84  -135 ± 0.53
#> 10  -126 ± NA  -0.0104 ± 0.0033  -133 ± 0.81  -133 ± 0.72
#> # ℹ 10,388 more rows