Imported from llorracc/SolvingMicroDSOPs (AGENTS.md). Install upstream with npx skills add llorracc/SolvingMicroDSOPs. Copyright stays with the author.

AGENTS.md — Orientation for AI-Assisted Work

This repository contains SolvingMicroDSOPs: LaTeX lecture notes plus replication code (Python, Stata) for numerical methods in consumption/saving problems under uncertainty. The main artifact is the book PDF built from root-level .tex files.

The document teaches how to solve consumption-saving problems efficiently, progressing from brute-force value function iteration through the transformation trick (inverse marginal value), the Endogenous Grid Method (EGM), and a modular stage architecture for composing complex periods. It concludes with structural estimation matching model predictions to Survey of Consumer Finances data.

Quick orientation

What this document contributes

See CONTRIBUTION.md for a concise statement of what the document covers, who it is for, and how it relates to HARK and standard textbooks.

Critical rules

• Matsya: When the user says to "ask Matsya" (or equivalent), show the full output that Matsya produces—do not replace it with a summary you construct.
• docs/ is a reference copy. Do not edit it. All edits use root-level files only.
• Build: latexmk on root .tex (uses LuaLaTeX via .latexmkrc)
• Code mirrors math: Python identifiers parallel LaTeX macros (e.g. DiscFac ↔ \DiscFac, vEndPrd ↔ \vEndPrd)
• verbatimwrite pattern: Many equation files in Equations/ are generated at compile time by \begin{verbatimwrite} blocks inside _sectn-*.tex files. Always edit the equation in the section file, not the generated .tex fragment in Equations/.

Detailed documentation: .agents/

For deeper context—LaTeX structure, build system, macro conventions, code layout, and the %% CodeInfo: annotation scheme—see the .agents/ directory:

Read these before making substantial changes to LaTeX or code.