Agent Skills
› Prismer-AI/Prismer
› quant-analysis
quant-analysis
GitHub提供量化金融分析能力，涵盖投资组合优化、风险建模及时间序列计量经济学。支持通过Jupyter执行Python代码进行数据清洗、因子模型构建、VaR计算及蒙特卡洛模拟，辅助投资决策与学术研究。
Trigger Scenarios
用户询问投资组合分析或股票收益
需要构建风险模型或波动率模型
请求进行资产配置或因子投资优化
Install
npx skills add Prismer-AI/Prismer --skill quant-analysis -g -y
SKILL.md
Frontmatter
{
"name": "quant-analysis",
"description": "Quantitative finance analysis including portfolio optimization, risk modeling, and time series econometrics using jupyter_execute. Use when the user asks about portfolio analysis, stock returns, financial risk, investment optimization, or volatility modeling."
}
Quantitative Analysis Skill
Description
Perform quantitative finance research including data analysis, portfolio optimization, risk modeling, and econometric analysis.
Tools Used
jupyter_execute- Execute Python code for financial analysis (auto-switches to Jupyter)jupyter_notebook- Manage analysis notebooksupdate_notebook- Set up analysis cells in Jupyterupdate_latex- Write finance paper content to LaTeX editorlatex_compile- Compile research papers (auto-switches to LaTeX editor)update_notes- Write analysis summaries and findings
Capabilities
Data Analysis
- Time series analysis of financial returns
- Cross-sectional regression (Fama-MacBeth, panel data)
- Event studies and abnormal return analysis
- Volatility modeling (GARCH family)
Portfolio Optimization
- Mean-variance optimization (Markowitz)
- Black-Litterman model with views
- Risk parity and equal risk contribution
- Factor-based portfolio construction
Risk Analysis
- Value-at-Risk (VaR) and Conditional VaR
- Stress testing and scenario analysis
- Copula-based dependency modeling
- Monte Carlo simulation
Usage Patterns
Analyze Returns
When user says: "Analyze the performance of [asset/portfolio]"
- Load price data using pandas/yfinance
- Calculate returns, volatility, Sharpe ratio
- Plot cumulative returns and drawdowns
- Run statistical tests (normality, autocorrelation)
- Present findings with charts
Build a Model
When user says: "Build a [pricing/risk/factor] model"
- Clarify model specification and data requirements
- Load and clean data
- Estimate model parameters
- Validate with out-of-sample testing
- Report results with diagnostics
Tool Examples
Load and analyze stock returns
# via jupyter_execute
import yfinance as yf
import pandas as pd
import numpy as np
data = yf.download("AAPL", start="2023-01-01", end="2024-01-01")
returns = data["Close"].pct_change().dropna()
print(f"Mean: {returns.mean():.4f}, Vol: {returns.std():.4f}, Sharpe: {returns.mean()/returns.std()*np.sqrt(252):.2f}")
Validation checkpoints
- Verify data has no missing values or extreme outliers before modeling
- Check model residuals for autocorrelation after estimation
- Confirm out-of-sample period has no look-ahead bias
Version History
- 2dbe71f Current 2026-08-20 10:22
