Add d-morrison/macros as git submodule with demo usage

.github/workflows/preview.yml

@@ -21,6 +21,7 @@ jobs:
       - name: Check out repository
         uses: actions/checkout@v4
         with:
+          submodules: true
           fetch-depth: 0  # Fetch full history for git diff
 
       - name: Set up Quarto

.github/workflows/publish.yml

@@ -13,6 +13,8 @@ jobs:
     steps:
       - name: Check out repository
         uses: actions/checkout@v4
+        with:
+          submodules: true
 
       - name: Set up Quarto
         uses: quarto-dev/quarto-actions/setup@v2

.gitignore

@@ -47,6 +47,7 @@ MANIFEST
 
 # TeX (LaTeX intermediate files)
 *.tex
+!macros-pdf.tex
 *.aux
 *.log
 *.toc

.gitmodules

@@ -0,0 +1,3 @@
+[submodule "macros"]
+	path = macros
+	url = https://github.com/d-morrison/macros.git

_quarto.yml

@@ -28,7 +28,11 @@ format:
     toc: true
     number-sections: true
     embed-resources: false
+    include-in-header:
+      - macros/macros.qmd
   pdf:
+    include-in-header:
+      - macros-pdf.tex
     geometry:
       - top=15mm
       - bottom=20mm

chapter2.qmd

@@ -14,6 +14,32 @@ $$
 \int_{-\infty}^{\infty} e^{-x^2} dx = \sqrt{\pi}
 $$
 
+## Custom Macros
+
+This book uses the [`d-morrison/macros`](https://github.com/d-morrison/macros) submodule to provide convenient LaTeX shorthand for statistical and mathematical notation.
+
+### Probability and Distributions
+
+The normal distribution $\Normal\paren{\m, \ss}$ can be written using macros as `$\Normal\paren{\m, \ss}$`, where `\m` expands to $\mu$ and `\ss` expands to $\sigma^2$.
+
+A random variable $X \dist \Normal\paren{\mu, \sigma^2}$ has expectation $\E{X} = \mu$ and variance $\Var{X} = \sigma^2$.
+
+### Regression Notation
+
+In linear regression, we estimate $\vbeta$ using ordinary least squares. The fitted values are $\hy = \mX \hb$, where $\hb = \inv{\mX' \mX} \mX' \vy$.
+
+The standard error of $\hb$ is $\hse{\hb} = \hs \sqrt{\inv{\mX' \mX}}$.
+
+### Likelihood and Estimation
+
+The log-likelihood function $\llik(\th)$ is maximized at the MLE $\hth_{\text{ML}}$.
+
+The score function is $\score(\th) = \deriv{\th} \llik(\th)$ and the observed information is $\oinf(\th) = -\hess(\th)$.
+
+### Logistic Regression
+
+The logit link function is $\logitf{\pi} = \logt\paren{\frac{\pi}{1 - \pi}}$ and its inverse is $\expitf{\eta} = \frac{e^\eta}{1 + e^\eta}$.
+
 ## Tables
 
 | Column 1 | Column 2 | Column 3 |