From 51b2f2711d795dba47865cb81bbcdf16edd8f1e0 Mon Sep 17 00:00:00 2001
From: Grant McDermott <grant.mcdermott@gmail.com>
Date: Mon, 16 Dec 2024 13:36:11 -0800
Subject: [PATCH] Lean false (#58)

* lean = FALSE

* gof catch

* news
---
 NEWS.md             |  8 +++++---
 R/emfx.R            | 17 ++++++++---------
 man/emfx.Rd         | 17 ++++++++---------
 vignettes/etwfe.Rmd |  1 +
 4 files changed, 22 insertions(+), 21 deletions(-)

diff --git a/NEWS.md b/NEWS.md
index b8d0645..ad7c2c8 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -7,9 +7,11 @@
 allows for obtaining the linear prediction for non-linear models. Internally
 passed as `marginaleffects::slopes(..., type = predict)`, thus avoiding a clash
 with the topline `emfx(..., type = <aggregration_type>)` argument. (#49)
-- `emfx(..., lean = TRUE)` ensures a light return object by default, by
-stripping away data-heavy attributes that are unlikely to be needed afterwards.
-(#51)
+- `emfx(..., lean = <logical>)`. Default value is `FALSE`, but switching to
+`TRUE` will ensure a light return object that strips away data-heavy attributes
+(e.g., copies of the original model). These attributes are unlikely to be needed
+as part of the `emfx()` workflow, so we may change the default to `lean = TRUE`
+in a future version of **etwfe**. (#51, #58)
 - Native `plot.emfx()` method (via a **tinyplot** backend) for visualizing
 `emfx` objects. (#54)
 
diff --git a/R/emfx.R b/R/emfx.R
index b54d0bf..b8d12da 100644
--- a/R/emfx.R
+++ b/R/emfx.R
@@ -51,14 +51,13 @@
 #'   that behaviour via `FALSE` is _strictly_ performative: the "zero" treatment
 #'   effects for any pre-treatment periods is purely an artefact of the
 #'   estimation setup.
-#' @param lean Logical. Enforces a lean return object; namely a simple
-#'   data.frame of the main results, stripped of ancillary attributes. Defaults
-#'   to `TRUE`, in which case `options(marginaleffects_lean = TRUE)` is set
-#'   internally at the start of the `emfx` call, before being reverted upon
-#'   exit. Note that this will disable some advanced `marginaleffects`
-#'   post-processing features, but those are unlikely to be used in the `emfx`
-#'   context and means that we can dramatically reduce the size of the return
-#'   object.
+#' @param lean Logical. Default is `FALSE`. Switching to `TRUE` enforces a lean
+#'   return object; namely a simple data.frame of the main results, stripped of
+#'   ancillary attributes. Note that this will disable some advanced
+#'   `marginaleffects` post-processing features, but those are unlikely to be
+#'   used in the `emfx` context. The upside is a potentially dramatic reduction
+#'   in the size of the return object. Consequently, we may change the default
+#'   to `TRUE` in a future version of **etwfe**.
 #' @param ... Additional arguments passed to
 #'   [`marginaleffects::slopes`]. For example, you can pass `vcov =
 #'   FALSE` to dramatically speed up estimation times of the main marginal
@@ -105,7 +104,7 @@ emfx = function(
     collapse = compress,
     predict = c("response", "link"),
     post_only = TRUE,
-    lean = TRUE,
+    lean = FALSE,
     ...
 ) {
   
diff --git a/man/emfx.Rd b/man/emfx.Rd
index 54620ee..390af7d 100644
--- a/man/emfx.Rd
+++ b/man/emfx.Rd
@@ -12,7 +12,7 @@ emfx(
   collapse = compress,
   predict = c("response", "link"),
   post_only = TRUE,
-  lean = TRUE,
+  lean = FALSE,
   ...
 )
 }
@@ -67,14 +67,13 @@ that behaviour via \code{FALSE} is \emph{strictly} performative: the "zero" trea
 effects for any pre-treatment periods is purely an artefact of the
 estimation setup.}
 
-\item{lean}{Logical. Enforces a lean return object; namely a simple
-data.frame of the main results, stripped of ancillary attributes. Defaults
-to \code{TRUE}, in which case \code{options(marginaleffects_lean = TRUE)} is set
-internally at the start of the \code{emfx} call, before being reverted upon
-exit. Note that this will disable some advanced \code{marginaleffects}
-post-processing features, but those are unlikely to be used in the \code{emfx}
-context and means that we can dramatically reduce the size of the return
-object.}
+\item{lean}{Logical. Default is \code{FALSE}. Switching to \code{TRUE} enforces a lean
+return object; namely a simple data.frame of the main results, stripped of
+ancillary attributes. Note that this will disable some advanced
+\code{marginaleffects} post-processing features, but those are unlikely to be
+used in the \code{emfx} context. The upside is a potentially dramatic reduction
+in the size of the return object. Consequently, we may change the default
+to \code{TRUE} in a future version of \strong{etwfe}.}
 
 \item{...}{Additional arguments passed to
 \code{\link[marginaleffects:slopes]{marginaleffects::slopes}}. For example, you can pass \code{vcov = FALSE} to dramatically speed up estimation times of the main marginal
diff --git a/vignettes/etwfe.Rmd b/vignettes/etwfe.Rmd
index 913cf0a..73f12e6 100644
--- a/vignettes/etwfe.Rmd
+++ b/vignettes/etwfe.Rmd
@@ -317,6 +317,7 @@ modelsummary(
   list(mod_es2, mod_es),
   shape       = term:event:statistic ~ model,
   coef_rename = rename_fn,
+  gof_omit    = "Adj|Within|IC|RMSE",
   stars       = TRUE,
   title       = "Event study",
   notes       = "Std. errors are clustered at the county level"