Apply suggestions from code review Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

.Rbuildignore

@@ -13,3 +13,5 @@ _freeze/
 ^\.github$
 ^README\.Rmd$
 ^data-raw$
+^\.positai$
+^\.claude$

.github/copilot-instructions.md

@@ -188,6 +188,7 @@ Key macros to use:
 - **Aligned equations**: Use `\ba` / `\ea` for `\begin{aligned}` / `\end{aligned}`
 - **Greek letters**: Use `\b` for $\beta$, `\g` for $\gamma$, `\a` for $\alpha$
 - **Formatting**: Use `\red{...}` and `\blue{...}` for colored text in math
+- **Deviation/error notation**: Use `\erf{...}` for deviations of estimates/estimators from their estimands; use `\devn(...)` for all other deviations (e.g., observations from population means)
 
 matrix-product helper macros:
 
@@ -198,7 +199,7 @@ matrix-product helper macros:
 
 Residual and deviation helper macros include:
 
-- `\err` for generic error or residual terms
+- `\err` is deprecated; prefer `\devn(...)` for new non-estimation deviations while legacy uses are being phased out
 - `\erf{\theta}` for estimate/estimand deviations
 - `\devn(...)` for other deviations
 - `\resid` for residual symbols (`r`)
@@ -213,6 +214,11 @@ Residual and deviation helper macros include:
 
 Always check `latex-macros/macros.qmd` for available macros before writing raw LaTeX.
 
+When a repeatedly used expression needs a new macro,
+add it to `latex-macros/macros.qmd`
+in the `latex-macros` submodule
+and push that submodule update.
+
 ## Color Coding Strategy for Math Expressions
 
 Use `\red{...}` and `\blue{...}` purposefully and consistently to help readers:

.github/workflows/lint-changed-files.yaml

@@ -16,13 +16,18 @@ jobs:
       - uses: actions/checkout@v5
 
       - uses: r-lib/actions/setup-r@v2
+        with:
+          r-version: '4.6.0'
+          use-public-rspm: true
 
       - name: Install system dependencies
         run: |
           sudo apt-get update
-          sudo apt-get install -y jags libcurl4-openssl-dev libpng-dev libfontconfig1-dev libjpeg-dev libglpk-dev libharfbuzz-dev libfribidi-dev libfreetype6-dev libtiff5-dev libwebp-dev
+          sudo apt-get install -y jags libcurl4-openssl-dev libpng-dev libfontconfig1-dev libjpeg-dev libglpk-dev libharfbuzz-dev libfribidi-dev libfreetype6-dev libtiff5-dev libwebp-dev libnode109
 
       - uses: r-lib/actions/setup-renv@v2
+        env:
+          GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
 
       - name: Add lintr options
         run: |
@@ -33,6 +38,7 @@ jobs:
         run: |
           files <- gh::gh("GET https://api.github.com/repos/${{ github.repository }}/pulls/${{ github.event.pull_request.number }}/files")
           changed_files <- purrr::map_chr(files, "filename")
+          changed_files <- setdiff(changed_files, "renv/activate.R")
           all_files <- list.files(recursive = TRUE)
           exclusions_list <- as.list(setdiff(all_files, changed_files))
           lintr::lint_dir(exclusions = exclusions_list)

.github/workflows/preview.yml

@@ -43,18 +43,21 @@ jobs:
         with:
           tinytex: false
 
-      - uses: r-lib/actions/setup-r@HEAD
+      - uses: r-lib/actions/setup-r@v2
         with:
+          r-version: '4.6.0'
           use-public-rspm: true
 
       - name: Install system dependencies
         run: |
           sudo apt-get update
           # `libglpk-dev` is required for the igraph/ggdag/ggraph dependency stack.
           # Keep this dependency aligned with other workflows that restore/render the same renv set.
-          sudo apt-get install -y jags libcurl4-openssl-dev libpng-dev libfontconfig1-dev libjpeg-dev libglpk-dev libharfbuzz-dev libfribidi-dev libfreetype6-dev libtiff5-dev libwebp-dev
+          sudo apt-get install -y jags libcurl4-openssl-dev libpng-dev libfontconfig1-dev libjpeg-dev libglpk-dev libharfbuzz-dev libfribidi-dev libfreetype6-dev libtiff5-dev libwebp-dev libnode109
 
       - uses: r-lib/actions/setup-renv@v2
+        env:
+          GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
 
       - name: Install local package
         run: |

.github/workflows/publish.yml

@@ -28,16 +28,19 @@ jobs:
         with:
           tinytex: true
 
-      - uses: r-lib/actions/setup-r@HEAD
+      - uses: r-lib/actions/setup-r@v2
         with:
+          r-version: '4.6.0'
           use-public-rspm: true
 
       - name: Install system dependencies
         run: |
           sudo apt-get update
-          sudo apt-get install -y jags libcurl4-openssl-dev libpng-dev libfontconfig1-dev libjpeg-dev libglpk-dev libharfbuzz-dev libfribidi-dev libfreetype6-dev libtiff5-dev libwebp-dev
+          sudo apt-get install -y jags libcurl4-openssl-dev libpng-dev libfontconfig1-dev libjpeg-dev libglpk-dev libharfbuzz-dev libfribidi-dev libfreetype6-dev libtiff5-dev libwebp-dev libnode109
 
       - uses: r-lib/actions/setup-renv@v2
+        env:
+          GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
 
       - name: Install local package
         run: |

.gitignore

@@ -43,3 +43,5 @@ renv/lock/
 renv/python/
 renv/sandbox/
 renv/staging/
+.positai
+.claude

DESCRIPTION

@@ -53,6 +53,7 @@ Suggests:
     plotly,
     rjags,
     runjags,
+    datawizard (>= 1.3.1),
     survminer,
     tidyr,
     spelling,

_subfiles/Linear-models-overview/_def-residual-deviation.qmd

@@ -1,26 +1,59 @@
-:::{#def-resid-noise}
-#### Residual noise/deviation from the population mean
+:::{#def-deviation-subpop-mean}
+#### Deviation of an observation from its subpopulation mean
 
-The **residual noise** in a probabilistic model $p(Y)$,
-also known as the 
-**residual deviation of an observation from its population mean**
-or **residual** for short,
-is the difference between an observed value $y$ and its population mean:
+The **deviation of an observation from its subpopulation mean**
+in a probabilistic model $p(Y \mid X)$,
+is the difference between an observed value
+and its model-implied mean given covariates:
 
-$$\devn(y) \eqdef y - \Expf{Y}$$ {#eq-def-resid}
+$$\devn(y_i) \eqdef y_i - \E{Y_i \mid X_i}$$ {#eq-def-deviation-subpop-mean}
 :::
 
 :::{.notes}
-We use the same notation for residual noise that we used for [errors](estimation.qmd#def-error).
+Many sources call deviations "error" or "noise."
+The model-implied mean
+can be viewed as an estimate of $Y_i$,
+before $y_i$ is observed.
+However, an estimation error is defined as an estimate minus its estimand
+(see [estimation error](estimation.qmd#def-estimation-error)),
+and deviation is the observed value minus its estimand,
+so the deviation is actually the *negative* of the estimation error of the mean $\E{Y_i \mid X_i}$
+with respect to its estimand $Y_i$:
 
-$\Expf{Y}$ can be viewed as an estimate of $Y$, before $y$ is observed.
-Conversely, each observation $y$ can be viewed as an estimate of $\Expf{Y}$
-(albeit an imprecise one, individually, since $n=1$).
+$$\devn(y_i) = -\paren{\E{Y_i \mid X_i} - y_i}$$
+
+On the other hand,
+each observation $y_i$ can be viewed
+as a nonparametric estimate of $\mu_i = \E{Y_i \mid X_i}$
+(albeit an imprecise one, individually, since $n=1$):
+
+$$y_i = \hmu_i^{(NP)}$$
+
+where $\hmu_i^{(NP)}$ denotes
+the nonparametric estimate from a single observed value $y_i$.
+
+Thus,
+the deviation can be interpreted as the estimation error of $y_i$
+with respect to $\E{Y_i \mid X_i}$:
+
+$$
+\ba
+\devn(y_i)
+&= y_i - \mu_i
+\\
+&= \hmu_i^{(NP)} - \mu_i
+\\
+&= \erf{\hmu_i^{(NP)}}
+\ea
+$$
 
 :::
 
-We can rearrange @eq-def-resid
-to view $y$ as
-the sum of its mean plus the residual noise:
+---
+
+We can rearrange @eq-def-deviation-subpop-mean
+to view $y_i$ as
+the sum of its mean plus the deviation
+(often called error/noise):
 
-$$y = \Exp{Y} + \epsf{y}$$
+$$y_i = \E{Y_i \mid X_i} + \devn(y_i)$$

_subfiles/Linear-models-overview/_def-signal.qmd

@@ -0,0 +1,9 @@
+:::{#def-signal}
+#### Signal (statistical sense)
+
+In statistical modeling,
+the **signal** is the deterministic part of the model.
+For mean-based models,
+the signal is the model-implied mean function,
+for example $\E{Y \mid X}$ (or $\E{Y}$ when there are no covariates).
+:::