Compare one R quantity against its SAS reference

Errors – never warns, never skips – when the quantity is absent on either side. A comparison that cannot fail is worse than no comparison.

Usage

compare_parity(quantity, r, sas, class, source = "lst", digits = NA_integer_)

Arguments

quantity

Name of the quantity, used in the report.

r

Value computed in R.

sas

Value from the SAS reference.

class

Tolerance class; see parity_tolerance.

source

Where the SAS value came from – "lst" or "outhaz".

digits

For class = "printed", the number of decimal places the reference was printed to, as a single non-negative whole number. The tolerance becomes half of the last place.

Value

A one-row data frame.

Details

The outcome is three-state. R_BETTER fires only for a log-likelihood that exceeds SAS's beyond tolerance: a multi-start optimizer regularly beats a single-start one, and recording that as a failure would train the reader to distrust a real improvement.

Relative discrepancy at a zero reference

When the SAS value is zero there is no relative scale. The reported rel_diff is 0 if the two agree exactly and Inf otherwise – never NA, which parity_headline would drop, reporting a real failure as no discrepancy at all.

Examples

compare_parity("log_likelihood", r = -239.1941, sas = -239.194,
               class = "loglik")
#>         quantity         r      sas source abs_diff     rel_diff rtol  atol
#> 1 log_likelihood -239.1941 -239.194    lst    1e-04 4.180707e-07    0 5e-04
#>   outcome
#> 1    PASS