Chaos-Hyper/dsh-econ-tools
dsh-econ-tools
Econometrics Research Assistant — A DSH plugin with 6 tools for econometrics analysis
Install
npx @deepseek-ai/dsh plugin --profile web add github:Chaos-Hyper/dsh-econ-toolsRestart `dsh web` after install. Bundle APIs can change during the developer preview.
README badge
[](https://dshhub.dev/plugins/dsh-econ-tools)Paste this into your README. The star count updates with every catalog sync.
From the README
Excerpt from Chaos-Hyper/dsh-econ-tools, cleaned of badges and images.
📊 dsh-econ-tools — Econometrics Research Assistant
English | 中文
A DeepSeek Harness plugin providing 6 ready-to-use econometrics tools covering the full research workflow: method selection, data preparation, model specification, empirical analysis, robustness checks, and result reporting.
Feature Overview
| Tool | Function | Use Case |
|---|---|---|
🎯 econ_method_guide | Method Guide — Recommend appropriate econometric models based on research question and data type | Research design stage, unsure which model to use |
🧹 econ_data_prep | Data Preparation — Missing value handling, outlier detection, variable transformation, categorical encoding, with Python code snippets | Cleaning raw data before analysis |
⚙️ econ_model_spec | Model Specification & Variable Selection — Theory-driven, data-driven, hybrid, and ML-based (LASSO/Ridge/ElasticNet) strategies, with diagnostic checklists | Selecting core variables and controls |
🔬 econ_run_analysis | Empirical Analysis — Supports OLS, IV/2SLS, Logit, Probit, panel FE, DID, RDD; auto-generates Python/R/Stata code templates with interpretation guidance | Running regressions, interpreting results |
🛡️ econ_robustness | Robustness Checks — Omitted variables, measurement error, sample selection, model specification, outliers, parallel trends, placebo tests — 7 dimensions | Verifying whether core findings are reliable |
📝 econ_report | Result Reporting — Generate descriptive statistics tables, baseline regression tables, and robustness check summaries in Markdown / LaTeX / HTML, bilingual (CN/EN) | Writing papers, formatting result tables |
Quick Start
Installation
Option 1: From GitHub (Recommended)
Install directly via the dsh CLI:
dsh plugin --profile web add github:Chaos-Hyper/dsh-econ-tools
Option 2: Local File Installation
If you already have the source directory, install by path:
dsh plugin --profile web add /path/to/dsh-econ-tools
Or manually add it to the web profile dependencies (edit ~/.dsh/profiles/web/package.json):
"dependencies": {
"dsh-econ-tools": "link:/path/to/dsh-econ-tools"
}
Then add "dsh-econ-tools" to the dsh.profile.bundles array, and run:
cd ~/.dsh/profiles/web
pnpm install
Restart DSH for the changes to take effect.
Usage
The Agent will automatically call the appropriate tool based on your research needs. For example:
"I want to study the impact of education on income using cross-sectional data. What model should I use?" → Agent calls
econ_method_guide, recommending OLS, IV methods, etc.
"Run robustness checks for potential omitted variable bias." → Agent calls
econ_robustness, providing Oster stability test and other solutions.
Tool Details
1. econ_method_guide
Parameters:
research_goal: Research goal (causal inference / prediction / policy evaluation)dependent_type: Dependent variable type (continuous / binary / panel)data_structure: Data structure (cross-section / time series / panel)endogeneity_concern: Whether endogeneity is a concern (optional)
Sample output:
{
"recommended_models": ["OLS", "DID"],
"methodology_notes": ["Run model diagnostics", "Use robust standard errors"],
"next_tools": ["econ_data_prep", "econ_model_spec", "econ_run_analysis"]
}
2. econ_data_prep
…
