Add induced_priors lecture: induced priors, VARs, and sign restrictions

[jstac]

Adds a new lecture, induced_priors, to the advanced series (after supply_demand_var), with the associated bibliography entries.

Written by @thomassargent30 (commit a457ce8 on the uhlig branch). Opened as a PR so the lecture gets a preview build and review comments.

Changes:

• lectures/induced_priors.md (new)
• lectures/_toc.yml: add induced_priors
• lectures/_static/quant-econ.bib: six new entries

🤖 Generated with Claude Code

[netlify]

✅ Deploy Preview for lustrous-melomakarona-3ee73e ready!

Name	Link
🔨 Latest commit	10ee640
🔍 Latest deploy log	https://app.netlify.com/projects/lustrous-melomakarona-3ee73e/deploys/6acd701ede5b3d67552f1afa
😎 Deploy Preview	https://deploy-preview-390--lustrous-melomakarona-3ee73e.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify project configuration.

[jstac]

First-round review: style and typos

This is a basic pass against the QuantEcon style guide (manual: styleguide/). It does not review the economics or the econometrics. Line numbers refer to lectures/induced_priors.md at a457ce8.

Typos and wording

1. L50–56 (overview list). Both the fourth and the fifth items end in "and". Drop the "and" after "structural impulse responses" (L51).
2. L387, L447, L1473. var_names[i][:5] cuts "quantity $q$" down to "quant", so the printed output reads impact of supply on quant : .... A separate short list such as short_names = ["q", "p"] would fix this.
3. L1567. "The widest prior on $g_d$, $U[0.005, 1]$, also spans a factor of 200". The $g_s$ range spans a factor of about 27, so "also" reads oddly. Suggest: "... spans an even wider range, a factor of 200."
4. L1615. "for a given $g_s$, the green region extends to larger values of $g_s$ when $g_d$ is large". The opening clause conflicts with the rest of the sentence. Suggest dropping "for a given $g_s$,".
5. L1489. "Uhlig's Figure 7" needs a citation: "Figure 7 of {cite:t}Uhlig2017".
6. L1367. "Exercise 2 of {doc}supply_demand_var" could link directly to the exercise: "{ref}sdvar_ex2 in {doc}supply_demand_var".

Citations

In several places the author's name is part of the sentence but the citation uses the parenthetical {cite}. This renders as "Uhlig [Uhlig, 2005]". The style guide asks for {cite:t} here:

Line	Current	Suggested
L53	in the spirit of Harald Uhlig {cite}`Uhlig2005,Uhlig2017`	in the spirit of {cite:t}`Uhlig2005,Uhlig2017`
L87	Uhlig {cite}`Uhlig2005` proposed	{cite:t}`Uhlig2005` proposed
L114	In {cite}`Uhlig2017` he illustrates	{cite:t}`Uhlig2017` illustrates
L298	Geweke {cite}`Geweke2010` calls	{cite:t}`Geweke2010` calls
L300, L498	Leeper, Traum, and Walker {cite}`LeeperTraumWalker2017`	{cite:t}`LeeperTraumWalker2017`
L306	As Uhlig {cite}`Uhlig2017` (his Principle 17) reminds us	As {cite:t}`Uhlig2017` reminds us (his Principle 17)
L561–562	Del Negro and Schorfheide {cite}`DelNegroSchorfheide2004`	{cite:t}`DelNegroSchorfheide2004`
L614, L936	Uhlig {cite}`Uhlig2017`	{cite:t}`Uhlig2017`
L867	Figure 2 of {cite}`Uhlig2005`	Figure 2 of {cite:t}`Uhlig2005`
L1321	Baumeister and Hamilton {cite}`BaumeisterHamilton2015`	{cite:t}`BaumeisterHamilton2015`

Forms such as "Uhlig's Proposition A.1 {cite}`Uhlig2005`" (L760, L1444) are fine as they are.

Bibliography (quant-econ.bib)

The six new entries were checked against Crossref. Authors, years, volumes and pages all match. To meet the manual's bibliography rules:

• DOIs are missing from every entry (rule 9). They are:
  • Uhlig2005: 10.1016/j.jmoneco.2004.05.007
  • Uhlig2017: 10.1017/9781108227223.004
  • DelNegroSchorfheide2004: 10.1111/j.1468-2354.2004.00139.x
  • BaumeisterHamilton2015: 10.3982/ECTA12356
  • LeeperTraumWalker2017: 10.1257/aer.20111196
  • Geweke2010: 10.1515/9781400835249
• Uhlig2005 title is in sentence case. Rule 4 asks for title case: What Are the Effects of Monetary Policy on Output? Results from an Agnostic Identification Procedure.
• Uhlig2017 editor: write Honoré directly rather than Honor{\'e} (rule 11).

Math notation

• Expectation. The manual uses \mathbb{E}. Plain E appears at L89, L314, L316 and L553. (supply_demand_var also uses plain E, so this could be done for both lectures at once.)
• Probability. At L1313, $P(\text{fundamental})$ should be written $\mathbb{P}{\ldots}$.
• Overloaded symbols (worth considering).
  • $L$ is both the lag operator in $a(L)$, $B_s(L)$, $B_d(L)$ and the Cholesky factor in $L$, $L_q$, $|L^{-1} d|$.
  • The letter $a$ is used for the VAR innovation $a_t$, the polynomial $a(L)$ and Uhlig's impulse vector $a$ (L102).
  • A different symbol for the Cholesky factor would remove the clash that matters most.
• Leading decimals such as .8L (L161, L165, L421) could be written 0.8L. Optional, since the companion lecture uses the same form.

Figures

• Captions. No main-text figure has mystnb caption and name metadata. The manual asks for both on every code-generated figure outside exercises and solutions. Figures inside solutions are correctly left uncaptioned.
• Titles on single-panel figures (L573, L1069, L1149, L1232, L1296). The manual reserves ax.set_title for subplots, so these titles would move into the caption. The L1069 title is also long.
• Line width. The true-parameter and Cholesky lines use lw=1.5 (L404, L881). The manual asks for lw=2.
• Figure size. figsize is set on every figure. That makes sense for the multi-panel figures. The single-panel ones (L569, L1064, L1147, L1223, L1294) could use the default.

Code (minor)

• Several statements on one line joined by ; (L230, L231, L234, L236, L237). Splitting them would follow PEP8.
• L914. Use parentheses instead of the backslash continuation.
• L1469. rng_s reads like a random number generator; bounds would be clearer.
• L81. from quantecon import LQ is redundant given import quantecon as qe; qe.LQ can be used instead.

Numbers quoted in the text

Some numbers in the prose depend on the random draws:

• "roughly nine times" (L735)
• "about $0.83$" (L790)
• "almost nine times" (L901)
• "roughly one quarter ... about 60 percent" (L1319, L1354)

They are worth checking against the outputs of the preview build once it finishes.

[github-actions]

🚀 Deployed on https://6acd7220060f0169c5854357--lustrous-melomakarona-3ee73e.netlify.app