Overview
This skill focuses onmulti-dimensional public mutual fund screening. By integrating data such as performance, risk control, fund manager capability, asset allocation, and industry preferences, it helps users quickly and precisely identify targets that match specific investment logic from a vast number of public mutual funds.
Data Toolset (gildata-aidata)
| Tool Name | Core Function | Key Parameters |
|---|---|---|
FundMultipleFactorFilter |
Intelligent Fund Selection | Accepts natural language queries and returns a fund list that meets multi-condition screening |
Core Screening Dimensions
1. Performance and Return Metrics
- Period Returns: Returns over the past 1 month, past 3 months, past 6 months, past 1 year, past 3 years, and year-to-date.
- Peer Ranking: Return ranking among同类基金 (e.g., "top 10% ranking").
- Excess Return: The extent by which it exceeds the performance benchmark.
- Historical Win Rate: The probability of profitability under different holding periods.
2. Risk and Drawdown Control
- Maximum Drawdown: Historical maximum drawdown magnitude (e.g., "maximum drawdown less than 20%").
- Sharpe Ratio: Risk-adjusted return metric (e.g., "Sharpe ratio greater than 1").
- Volatility: The intensity of net value fluctuations.
- Calmar Ratio: The ratio of annualized return to maximum drawdown.
3. Fund Manager Dimension
- Years of Experience: The fund manager's years managing the fund or total years of experience (e.g., "more than 5 years of experience").
- Assets Under Management: The total scale currently managed by the fund manager.
- Investment Style: Value, growth, balanced, large-cap/small-cap style.
- Signature Work: Whether there is a representative product with excellent long-term performance.
4. Fund Type and Asset Allocation
- Fund Classification: Equity, hybrid, bond, index, QDII, FOF, money market, etc.
- Asset Distribution: Stock position, bond position, cash ratio.
- Industry Preference: Heavily weighted industries (e.g., "main holdings in new energy, semiconductors").
- Holding Concentration: The proportion of the top ten heavily weighted stocks in net value.
5. Scale and Fees
- Fund Scale: The latest asset scale (e.g., "scale greater than 1 billion").
- Fee Structure: Management fee, custody fee, subscription and redemption fees.
Workflow
- Intent Recognition: Extract screening conditions from the user's natural language (e.g., "hybrid funds with top 10% returns over the past year and maximum drawdown less than 15%").
- Build Query: Combine the extracted conditions into a natural language query statement.
- Call Tool: Call
FundMultipleFactorFilterto get the screening results. - Result Parsing: Parse the returned fund list and extract key metrics (code, name, type, latest net value, period return, drawdown, Sharpe ratio, fund manager, etc.).
- Generate Report: Output a structured fund selection results table, along with a brief investment logic analysis.
Report Output Template
# 公募基金智能选基结果报告
## 筛选条件
* **[条件 1]**：[如 近1年收益排名前 10%]
* **[条件 2]**：[如 最大回撤 < 15%]
* **[条件 3]**：[如 混合型基金]
## 筛选结果 (Top N)
| 基金代码 | 基金名称 | 基金类型 | 最新净值 | 近1年收益 (%) | 最大回撤 (%) | 夏普比率 | 基金经理 | 基金规模 (亿) |
| :--- | :--- | :--- | :--- | :--- | :--- | :--- | :--- | :--- |
| 000001 | 示例混合 | 混合型 | 1.500 | 25.30 | -12.50 | 1.20 | 张三 | 50.2 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... |
## 投资逻辑点评
* **业绩特征**：[简述筛选结果的整体业绩表现]
* **风控能力**：[分析回撤控制及夏普比率表现]
* **经理风格**：[简述基金经理的投资风格及稳定性]
* **配置建议**：[基于筛选结果的简要策略提示]
* **风险提示**：[提醒用户注意市场风险及历史业绩不代表未来表现]
Usage Examples
- "Help me screen for **equity-biased mixed funds ranked in the top 20% for the past one-year return and with a maximum drawdown of less than 20%**."
- "Findlong-term high-performing funds whose fund managers have more than 8 years of experience and a Sharpe ratio greater than 1."
- "Are there anyactively managed equity funds with assets under management greater than 5 billion and major holdings in the consumer sector?"
- "Screen forfixed income+ funds with positive returns year to date and low volatility."