Skip to content

[information_transfer] Add a lecture on Green's incentives for information transfer - #1119

Merged
mmcky merged 4 commits into
mainfrom
nov_penn5
Nov 10, 2026
Merged

mmcky merged 4 commits into
mainfrom
nov_penn5

Conversation

@mmcky

@mmcky mmcky commented Nov 10, 2026

Copy link
Copy Markdown
Contributor

Add a lecture on Green (1982), "Statistical Decision Theory Requiring Incentives for Information Transfer", with the comments by Postlewaite and Riley and the posterior characterization of Blackwell's order by Marschak and Miyasawa (1968).

The lecture solves the planner's problem as a linear program, extends Green's two-action theorem to any number of states, gives a complete proof of his result on binomial channels, and checks garblings with a martingale-transport feasibility problem. Exercises cover Riley's example with payments, Postlewaite's four cases, role reversal, and Marschak and Miyasawa's conditions.

Place the lecture after blackwell_kihlstrom, add short return links in blackwell_kihlstrom, information_market_equilibrium, pricing_information, opt_transport, and util_rand_resp, and add twelve bibliography entries.

…ation transfer

Add a lecture on Green (1982), "Statistical Decision Theory Requiring
Incentives for Information Transfer", with the comments by Postlewaite
and Riley and the posterior characterization of Blackwell's order by
Marschak and Miyasawa (1968).

The lecture solves the planner's problem as a linear program, extends
Green's two-action theorem to any number of states, gives a complete
proof of his result on binomial channels, and checks garblings with a
martingale-transport feasibility problem. Exercises cover Riley's
example with payments, Postlewaite's four cases, role reversal, and
Marschak and Miyasawa's conditions.

Place the lecture after blackwell_kihlstrom, add short return links in
blackwell_kihlstrom, information_market_equilibrium, pricing_information,
opt_transport, and util_rand_resp, and add twelve bibliography entries.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Copilot AI balanced review requested due to automatic review settings November 10, 2026 02:31

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Note

Copilot was unable to complete this review at the selected tier due to model policy or plan restrictions.

🟡 Changes recommended

A moderate correctness issue remains in lectures/information_transfer.md around line 317 where the code can return an inconsistent planner value versus the returned plan Z because planner is not recomputed after x is updated in the ties-handling LP.

1 open issue
What changed in this PR

Adds a new QuantEcon/MyST lecture on Green (1982) and incentive-compatible information transfer, connecting it to the existing information-economics sequence (Blackwell order, information markets, pricing/screening, and optimal transport), and wires it into the book structure with cross-links and citations.

Changes:

  • Introduces lectures/information_transfer.md with theory, LP formulations, garbling checks (martingale/transport feasibility), and exercises/solutions.
  • Updates navigation: inserts the new lecture after blackwell_kihlstrom in _toc.yml and adds short cross-references/return links in related lectures.
  • Extends lectures/_static/quant-econ.bib with the new references cited by the lecture.

Required changes from review comments (moderate):

  • In information_transfer.md (around line 317): when ties != None, recompute planner after updating x (or keep planner_opt separately) so the returned objective value matches the returned plan Z.
  • In information_transfer.md (around line 725): align garbling() with the rest of the lecture by disabling HiGHS presolve to avoid false infeasibility on small LP feasibility problems.
  • In information_transfer.md (around line 1225): disable HiGHS presolve in solve_with_pay() as well for consistency/robustness with other small LP solves.
File Reviewed change
lectures/​util_rand_resp.md Adds a brief cross-reference/return link relating informativeness/truthfulness to the new lecture.
lectures/​pricing_information.md Adds a short cross-reference contrasting pricing/screening with the information-transfer setting.
lectures/​opt_transport.md Adds a short cross-reference noting the transport/feasibility angle used for garbling checks.
lectures/​information_transfer.md Adds the new lecture content, code cells, examples, and exercises/solutions (with noted LP-solver consistency items to address).
lectures/​information_market_equilibrium.md Adds a short cross-link to the new lecture as a related information-channel model.
lectures/​blackwell_kihlstrom.md Adds forward reference/return link positioning the new lecture in the sequence.
lectures/​_toc.yml Inserts information_transfer after blackwell_kihlstrom in the book TOC.
lectures/​_static/​quant-econ.bib Adds the bibliography entries required by the new lecture and references.

This pull request requires changes.

Review effort: Balanced

Comment thread lectures/information_transfer.md Resolved
mmcky and others added 2 commits November 10, 2026 13:43
A figure caption inside a solution block makes the PDF build fail with
"Not in outer par mode", so plot Riley's example as a plain output.
Nothing referenced fig-it-riley.

Also add the page range to the Holmstrom1984 bibliography entry.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…ption

Style-guide pass: write the ten displayed matrices with bmatrix rather
than pmatrix (qe-math-003), and cut the fig-it-refine caption to five
words (qe-fig-004).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@github-actions

github-actions Bot commented Nov 10, 2026 •

Copy link
Copy Markdown

solve_planner and solve_agent_leads already switch presolve off; do the
same in garbling and solve_with_pay so all four solves are configured
alike. In garbling an infeasibility report is read as "no garbling
exists", so a spurious one would give a silently wrong answer. Outputs
are unchanged.

Addresses Copilot's review overview on this PR.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@mmcky

mmcky commented Nov 10, 2026

Copy link
Copy Markdown
Contributor Author

Addressed Copilot's review.

  • Inline comment on solve_planner (recompute planner after the tie-break): no change, deliberately. The returned value is the linear program's optimum, the tie-broken plan attains it to within the 1e-8 tolerance (measured at most 1.0e-8 relative over 4,800 solves), and payoffs() relies on it being the true optimum. Details are in the thread: [information_transfer] Add a lecture on Green's incentives for information transfer #1119 (comment)
  • Overview items on HiGHS presolve in garbling() and solve_with_pay(): done in dc5716a. All four linear programs now switch presolve off, as solve_planner and solve_agent_leads already did. In garbling() an infeasibility report is read as "no garbling exists", so a spurious one would otherwise give a silently wrong answer. The lecture's outputs are unchanged.

@mmcky

mmcky commented Nov 10, 2026

Copy link
Copy Markdown
Contributor Author

thanks @thomassargent30

@mmcky
mmcky merged commit a268b8f into main Nov 10, 2026
2 checks passed
@mmcky
mmcky deleted the nov_penn5 branch November 10, 2026 04:11
mmcky added a commit to QuantEcon/lectures that referenced this pull request Nov 10, 2026
…#50)

The nightly refreshes #48 and #49
left the drift check red on two counts; this settles both.

The bibliography. QuantEcon/lecture-python.myst#1113 merged
intermediate's two entries for Lucas and Prescott (1971) as
Lucas_Prescott_1971 and dropped LucasPrescott1971, so the ruling for
LucasPrescott1971 in sync/bib.yml, which took intermediate's copy, can
no longer apply and tools/bib-merge refused the key. Intro, advanced
and jax hold identical copies of the old entry, and no pool lecture
cites it since #48, so the ruling is removed rather
than replaced: the result is the entry `tools/bib-merge --propose`
proposes (take intro, the three seals equal). `tools/bib-merge --write`
regenerates lectures/_static/quant-econ.bib, whose LucasPrescott1971 is
now the three series' copy, normalised, without the volume, number and
DOI that only intermediate's copy had; the key the lectures cite,
Lucas_Prescott_1971, keeps them. 906 keys, 56 conflicts, 59 rulings,
nothing refused.

The toc. QuantEcon/lecture-python.myst#1110 and
QuantEcon/lecture-python.myst#1119 added lucas_prescott_search and
information_transfer to intermediate's toc, which drift issue
#47 reports. `tools/promote --toc intermediate
--exclude jax` promotes both, canonical in intermediate with no assets
and no rewrites, so the {doc}`information_transfer` references that
#49 brought into five pool lectures now resolve.
pandas_panel, unchanged in text, records intermediate's current pin,
a268b8f, for its second source. `tools/drift-check --accept-toc
intermediate` then puts intermediate's _toc.yml under the toc watch at
a268b8f, and the README's count of pool lectures follows, from 296 to
298.


Claude-Session: https://claude.ai/code/session_01WT8bMXuHMiWb8victibwcU

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
mmcky added a commit that referenced this pull request Nov 11, 2026
…ions (#1121)

sphinx-exercise 1.2.2 writes a captioned figure inside an exercise or solution box as `[H]`, as Sphinx does inside its own admonitions, so the PDF build no longer stops with "Not in outer par mode" (executablebooks/sphinx-exercise#94, executablebooks/sphinx-exercise#95).

- `environment.yml` and `.github/workflows/collab.yml`: sphinx-exercise 1.2.1 → 1.2.2.
- Restores the caption and name on nine figures in gated solutions, removed as the workaround in #1101, #1110 and #1119: exchange_rate_learning (2), genetic_classifier (1), learning_approximation (2), lucas_prescott_search (3), information_transfer (1).

Part of QuantEcon/workspace-lectures#70. The workaround is recorded on QuantEcon/meta#297.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants