From 003265587d1126ac51dc2460eb3f838248f8f502 Mon Sep 17 00:00:00 2001
From: Kyle Meyer <kylem@metrumrg.com>
Date: Tue, 30 Jan 2024 17:39:58 -0500
Subject: [PATCH] param docs: don't describe record names for priors as
 "informative"

The set-param and extract-param docs call records like THETAP and
THETAPV as "informative prior record names" because NONMEM refers to
records like THETAP and THETAPV as "informative record names".
However, that invites confusion with the unrelated statistical concept
of informative/non-informative priors (even more so because the nmrec
docs unnecessarily insert "prior" between "informative" and "record
names").

Drop "informative" entirely to avoid confusion.  While at it, reduce
"prior record names" to "record names" because "prior" makes it harder
to parse and is unnecessary given the context.

Thanks to @timwaterhouse for the suggestion.

Re: https://github.com/metrumresearchgroup/bbr/pull/635#issuecomment-1917912006
---
 R/extract-param.R    | 4 ++--
 R/set-param.R        | 4 ++--
 man/extract_param.Rd | 4 ++--
 man/set_param.Rd     | 2 +-
 4 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/R/extract-param.R b/R/extract-param.R
index 14fbbe4..386d2b0 100644
--- a/R/extract-param.R
+++ b/R/extract-param.R
@@ -20,8 +20,8 @@
 #'    example, a record of `$THETA (1)x4` is returned as `c(1, NA, NA, NA)`.
 #'
 #'  * If additional parameter records are used for priors, those are included in
-#'    the result. To avoid this, instead use informative prior record names
-#'    (such as `THETAP` and `THETAPV`).
+#'    the result. To avoid this, instead use more specific record names (such as
+#'    `THETAP` and `THETAPV`).
 #'
 #' @param records An [nmrec_ctl_records] object.
 #' @param mark_flags A vector of NONMEM flags (i.e. valueless options such as
diff --git a/R/set-param.R b/R/set-param.R
index f2d1b42..52aff6e 100644
--- a/R/set-param.R
+++ b/R/set-param.R
@@ -19,7 +19,7 @@
 #'
 #'  * Using additional parameter records for priors is not supported and will
 #'    lead to a size mismatch between the parameter and its records. Instead use
-#'    informative prior record names (such as `THETAP` and `THETAPV`).
+#'    more specific record names (such as `THETAP` and `THETAPV`).
 #'
 #' @param records An [nmrec_ctl_records] object.
 #' @param values Overwrite the parameter's initial estimates in `records` with
@@ -195,7 +195,7 @@ set_param <- function(records, name, values, fmt) {
         ),
         "i" = paste0(
           "If you're using ", toupper(name), " records for priors, ",
-          "please switch to informative record names ",
+          "please switch to more specific record names ",
           "(such as THETAP and THETAPV)."
         )
       ),
diff --git a/man/extract_param.Rd b/man/extract_param.Rd
index 906bf58..54ccd00 100644
--- a/man/extract_param.Rd
+++ b/man/extract_param.Rd
@@ -60,8 +60,8 @@ estimates for \code{OMEGA} and \code{SIGMA} records.
 defined in the control stream. All other values are returned as \code{NA}. For
 example, a record of \verb{$THETA (1)x4} is returned as \code{c(1, NA, NA, NA)}.
 \item If additional parameter records are used for priors, those are included in
-the result. To avoid this, instead use informative prior record names
-(such as \code{THETAP} and \code{THETAPV}).
+the result. To avoid this, instead use more specific record names (such as
+\code{THETAP} and \code{THETAPV}).
 }
 }
 }
diff --git a/man/set_param.Rd b/man/set_param.Rd
index 3e03f13..b43d3e0 100644
--- a/man/set_param.Rd
+++ b/man/set_param.Rd
@@ -72,7 +72,7 @@ initial estimates, but only the first explicitly appears. Calling
 by the parameter records.
 \item Using additional parameter records for priors is not supported and will
 lead to a size mismatch between the parameter and its records. Instead use
-informative prior record names (such as \code{THETAP} and \code{THETAPV}).
+more specific record names (such as \code{THETAP} and \code{THETAPV}).
 }
 }
 }