docs(readme): add causal inference coverage matrix + ecosystem stats page

Adds a 15-row coverage matrix at the top of README.md and README_CN.md
positioning StatsPAI against Stata (official + SSC), R (CRAN),
statsmodels+linearmodels, and DoubleML across the core causal-inference
families (DiD, IV, RD, Synth, DML, TMLE, neural causal, discovery,
Proximal, QTE, MR, Conformal, BCF, Spatial).

New docs/stats.md carries the full 23-row matrix, cross-ecosystem LOC
comparison (StatsPAI 188K vs statsmodels 382K vs Stata official 1.04M vs
R ecosystem ~150M), per-module LOC/function breakdown for all 78
modules, and a reproduction recipe. mkdocs nav wires it in as a
top-level "Ecosystem stats" entry.

LOC numbers are measured 2026-04-21 on macOS arm64; SSC / CRAN /
Bioconductor totals are flagged as estimates with sources.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: brycewang-stanford

README.md

@@ -15,6 +15,36 @@ StatsPAI is the **agent-native** Python package for causal inference and applied
 
 It brings R's [Causal Inference Task View](https://cran.r-project.org/web/views/CausalInference.html) (fixest, did, rdrobust, gsynth, DoubleML, MatchIt, CausalImpact, ...) and Stata's core econometrics commands into a single, consistent Python API.
 
+---
+
+## 📊 Causal Inference Coverage at a Glance
+
+StatsPAI's focus is **causal inference** — and on this axis we aim to be the most complete single package in any language. "Stata" = official + major SSC packages. "R" = CRAN. "sm+lm" = statsmodels + linearmodels.
+
+| Method family                                                 | Stata | R | sm+lm | DoubleML | **StatsPAI** |
+| ------------------------------------------------------------- | :---: | :---: | :---: | :---: | :---: |
+| DiD — staggered (CS/SA/BJS/dCdH/Gardner/Wooldridge ET) + event-study + honest CIs | ⚠️ | ✅ | ❌ | ❌ | 🏆 |
+| IV — classical (2SLS/LIML/GMM) + modern (Kernel IV / Deep IV / KAN-DeepIV) | ✅ classical only | ✅ classical only | ⚠️ classical | ⚠️ | 🏆 |
+| RD — CCT + 2D / boundary + multi-cutoff + honest CIs + ML-CATE (18+ estimators) | ⚠️ | ✅ (`rdrobust`) | ❌ | ❌ | 🏆 |
+| Synthetic Control — ADH / ASCM / gsynth / BSTS / Bayesian / PenSCM / FDID (20 methods) | ⚠️ | ⚠️ (7 pkgs) | ❌ | ❌ | 🏆 |
+| Double / Debiased ML                                          | ❌    | ✅   | ❌   | ✅   | ✅ |
+| Meta-Learners (S/T/X/R/DR) + Causal Forest / GRF              | ❌    | ✅   | ❌   | ❌   | ✅ |
+| TMLE / HAL-TMLE                                               | ❌    | ✅   | ❌   | ❌   | ✅ |
+| Neural causal (TARNet / CFRNet / DragonNet)                   | ❌    | ❌   | ❌   | ❌   | 🏆 |
+| Causal discovery (NOTEARS / PC / LiNGAM / GES)                | ❌    | ⚠️   | ❌   | ❌   | 🏆 |
+| Proximal CI (fortified / bidirectional / MTP / DNC)           | ❌    | ⚠️   | ❌   | ❌   | 🏆 |
+| QTE / distributional TE / CiC / dist-IV                       | ⚠️    | ⚠️   | ❌   | ❌   | ✅ |
+| Mendelian randomization (IVW/Egger/median/mode/PRESSO/MVMR/BMA) | ❌  | ✅   | ❌   | ❌   | ✅ |
+| Conformal causal inference                                    | ❌    | ❌   | ❌   | ❌   | 🏆 |
+| Bayesian Causal Forest (BCF / ordinal / factor-exposure)      | ❌    | ⚠️   | ❌   | ❌   | ✅ |
+| Spatial econometrics (weights → ESDA → ML/GMM → GWR/MGWR → panel) | ❌ | ⚠️ (5 pkgs) | ❌ | ❌ | 🏆 |
+
+**Legend**: 🏆 most complete across ecosystems · ✅ full coverage · ⚠️ partial / scattered / single algorithm · ❌ not available.
+
+**StatsPAI at a glance**: 889 registered functions · 78 modules · 188,244 LOC (core) + 42,768 LOC (tests). For the full coverage matrix (23 method families), per-module breakdown, and cross-ecosystem line-count comparison — see [`docs/stats.md`](docs/stats.md).
+
+---
+
 **🎉 NEW in v1.5.0 — Interference / Conformal / Mendelian family consolidation**
 
 StatsPAI 1.5.0 is a minor release bundling three concurrent improvements to the interference, conformal causal inference, and Mendelian Randomization families: full-family documentation guides, unified dispatchers matching the `sp.synth` / `sp.decompose` / `sp.dml` pattern, and a targeted correctness audit that fixed two silent-wrong-numbers issues.

README_CN.md

@@ -15,6 +15,36 @@ StatsPAI 是一个**面向 AI Agent** 的 Python 因果推断与应用计量经
 
 它将 R 的 [Causal Inference Task View](https://cran.r-project.org/web/views/CausalInference.html)（fixest、did、rdrobust、gsynth、DoubleML、MatchIt、CausalImpact、sfaR、lme4、oaxaca、ddecompose……）和 Stata 的核心计量命令（`frontier`、`xtfrontier`、`mixed`、`meglm`、`mixlogit`、`ivqreg`……），统一到一个一致的 Python API 中。
 
+---
+
+## 📊 因果推断覆盖一览
+
+StatsPAI 聚焦**因果推断**——在这条主线上，我们的目标是成为任何语言里覆盖最完整的单个包。"Stata" = 官方 + 主要 SSC 包；"R" = CRAN；"sm+lm" = statsmodels + linearmodels。
+
+| 方法家族                                                                 | Stata | R | sm+lm | DoubleML | **StatsPAI** |
+| ------------------------------------------------------------------------ | :---: | :---: | :---: | :---: | :---: |
+| DiD — 多期异质性（CS / SA / BJS / dCdH / Gardner / Wooldridge ET）+ 事件研究 + honest CIs | ⚠️ | ✅ | ❌ | ❌ | 🏆 |
+| IV — 经典（2SLS/LIML/GMM）+ 现代（Kernel IV / Deep IV / KAN-DeepIV）    | ✅ 仅经典 | ✅ 仅经典 | ⚠️ 仅经典 | ⚠️ | 🏆 |
+| RD — CCT + 2D / 边界 + 多断点 + honest CIs + ML-CATE（18+ 估计器）      | ⚠️ | ✅ (`rdrobust`) | ❌ | ❌ | 🏆 |
+| 合成控制 — ADH / ASCM / gsynth / BSTS / Bayesian / PenSCM / FDID（20 种方法） | ⚠️ | ⚠️ (7 个包) | ❌ | ❌ | 🏆 |
+| Double / Debiased ML                                                     | ❌    | ✅   | ❌    | ✅   | ✅ |
+| Meta-Learners（S/T/X/R/DR）+ 因果森林 / GRF                              | ❌    | ✅   | ❌    | ❌   | ✅ |
+| TMLE / HAL-TMLE                                                          | ❌    | ✅   | ❌    | ❌   | ✅ |
+| 神经因果（TARNet / CFRNet / DragonNet）                                  | ❌    | ❌   | ❌    | ❌   | 🏆 |
+| 因果发现（NOTEARS / PC / LiNGAM / GES）                                  | ❌    | ⚠️   | ❌    | ❌   | 🏆 |
+| Proximal 因果推断（fortified / bidirectional / MTP / DNC）               | ❌    | ⚠️   | ❌    | ❌   | 🏆 |
+| QTE / 分布处理效应 / CiC / dist-IV                                       | ⚠️    | ⚠️   | ❌    | ❌   | ✅ |
+| Mendelian randomization（IVW/Egger/median/mode/PRESSO/MVMR/BMA）         | ❌    | ✅   | ❌    | ❌   | ✅ |
+| Conformal 因果推断                                                       | ❌    | ❌   | ❌    | ❌   | 🏆 |
+| 贝叶斯因果森林（BCF / 有序 / 因子暴露）                                  | ❌    | ⚠️   | ❌    | ❌   | ✅ |
+| 空间计量（权重 → ESDA → ML/GMM → GWR/MGWR → 面板）                       | ❌    | ⚠️ (5 个包) | ❌ | ❌ | 🏆 |
+
+**图例**：🏆 跨生态最完整 · ✅ 完整覆盖 · ⚠️ 部分 / 分散 / 单算法 · ❌ 无。
+
+**StatsPAI 一句话概览**：889 个注册函数 · 78 个模块 · 188,244 行核心代码 + 42,768 行测试。完整覆盖矩阵（23 个方法家族）、按模块拆分、以及跨生态行数对比，详见 [`docs/stats.md`](docs/stats.md)。
+
+---
+
 **🎉 v1.5.0 新版本 — Interference / Conformal / Mendelian 三家族合并升级**
 
 StatsPAI 1.5.0 是 minor 版本，一次性完成 interference、conformal 因果推断、Mendelian 随机化三个家族的三项联动升级：完整家族文档指南、与 `sp.synth` / `sp.decompose` / `sp.dml` 同构的统一 dispatcher，以及一次针对性的正确性审计修掉两处"悄悄算错"。