fund-cli 2.0.0__py3-none-any.whl

fund_cli/analysis/holding.py

@@ -0,0 +1,217 @@
+"""
+持仓分析引擎
+
+提供基金持仓分析功能，包括行业配置、重仓股、集中度、变化追踪和风格分析。
+"""
+
+from typing import Any
+
+import numpy as np
+import pandas as pd
+
+from fund_cli.core.analyzer import Analyzer
+
+
+class HoldingAnalyzer(Analyzer):
+    """
+    持仓分析引擎
+
+    分析基金持仓数据，提供：
+    - 行业配置分析
+    - 重仓股分析
+    - 持仓集中度（HHI指数）
+    - 持仓变化追踪
+    - 风格分析（九宫格）
+    """
+
+    def analyze(self, data: pd.DataFrame, **kwargs: Any) -> dict[str, Any]:
+        """
+        执行综合持仓分析
+
+        Args:
+            data: 持仓数据 DataFrame，需包含 stock_code, stock_name, weight, industry 列
+
+        Returns:
+            综合分析结果字典
+        """
+        result = {}
+        if "industry" in data.columns:
+            result["industry_distribution"] = self.industry_distribution(data)
+        result["top_holdings"] = self.top_holdings(data)
+        result["concentration_hhi"] = self.concentration_hhi(data)
+        result["concentration_level"] = self._hhi_level(result["concentration_hhi"])
+        if "industry" in data.columns:
+            result["style_analysis"] = self.style_analysis(data)
+        return result
+
+    def industry_distribution(self, holdings: pd.DataFrame) -> dict[str, float]:
+        """
+        行业配置分析 (FUND-HOLDING-002)
+
+        Args:
+            holdings: 持仓数据，需包含 industry 和 weight 列
+
+        Returns:
+            行业分布字典 {行业名称: 占比(%)}
+
+        Raises:
+            ValueError: 数据缺少 industry 列
+        """
+        if "industry" not in holdings.columns:
+            raise ValueError("持仓数据缺少 industry 列")
+        distribution = holdings.groupby("industry")["weight"].sum()
+        return distribution.sort_values(ascending=False).to_dict()
+
+    def top_holdings(self, holdings: pd.DataFrame, top_n: int = 10) -> pd.DataFrame:
+        """
+        重仓股分析 (FUND-HOLDING-003)
+
+        Args:
+            holdings: 持仓数据
+            top_n: 返回前N大持仓，默认10
+
+        Returns:
+            前N大持仓 DataFrame
+        """
+        df = holdings.copy()
+        if "weight" in df.columns:
+            df = df.sort_values("weight", ascending=False)
+        return df.head(top_n).reset_index(drop=True)
+
+    def concentration_hhi(self, holdings: pd.DataFrame) -> float:
+        """
+        持仓集中度 - HHI指数 (FUND-HOLDING-004)
+
+        HHI = sum(weight_i^2)，其中 weight_i 为第i只股票占净值比例（小数形式）
+
+        Args:
+            holdings: 持仓数据，需包含 weight 列
+
+        Returns:
+            HHI指数值。>0.25 高度集中, 0.15~0.25 中度集中, <0.15 分散
+        """
+        if holdings.empty or "weight" not in holdings.columns:
+            return 0.0
+        weights = holdings["weight"].values / 100.0  # 转为小数
+        return float(np.sum(weights**2))
+
+    def track_changes(
+        self,
+        current: pd.DataFrame,
+        previous: pd.DataFrame,
+    ) -> pd.DataFrame:
+        """
+        持仓变化追踪 (FUND-HOLDING-005)
+
+        对比两期持仓数据，标识新增、删除、增持、减持的股票。
+
+        Args:
+            current: 当期持仓数据
+            previous: 上期持仓数据
+
+        Returns:
+            变化分析 DataFrame，包含 change_type 列（新增/删除/增持/减持/不变）
+        """
+        if current.empty or previous.empty:
+            return pd.DataFrame()
+
+        curr = (
+            current[["stock_code", "stock_name", "weight"]].copy()
+            if "stock_code" in current.columns
+            else current.copy()
+        )
+        prev = (
+            previous[["stock_code", "weight"]].copy()
+            if "stock_code" in previous.columns
+            else previous.copy()
+        )
+
+        merged = curr.merge(prev, on="stock_code", how="outer", suffixes=("_curr", "_prev"))
+
+        def classify_change(row: pd.Series) -> str:
+            if pd.isna(row.get("weight_prev")):
+                return "新增"
+            if pd.isna(row.get("weight_curr")):
+                return "删除"
+            diff = row["weight_curr"] - row["weight_prev"]
+            if abs(diff) < 0.01:
+                return "不变"
+            return "增持" if diff > 0 else "减持"
+
+        merged["change_type"] = merged.apply(classify_change, axis=1)
+        merged["weight_change"] = merged.get("weight_curr", 0) - merged.get("weight_prev", 0)
+        return merged.sort_values("weight_change", ascending=False, na_position="last").reset_index(
+            drop=True
+        )
+
+    def style_analysis(self, holdings: pd.DataFrame) -> dict[str, Any]:
+        """
+        风格分析 (FUND-HOLDING-006)
+
+        基于持仓股票的行业分布进行风格分析，输出风格九宫格位置。
+        简化实现：基于行业分布判断大盘/小盘和价值/成长倾向。
+
+        Args:
+            holdings: 持仓数据，需包含 industry 列
+
+        Returns:
+            风格分析结果字典
+        """
+        if "industry" not in holdings.columns:
+            return {"market_cap_style": "未知", "investment_style": "未知", "grid_position": "中"}
+
+        # 基于行业分布判断风格
+        distribution = self.industry_distribution(holdings)
+        total = sum(distribution.values()) or 1.0
+
+        # 大盘行业权重
+        large_cap_industries = {"银行", "非银金融", "食品饮料", "医药生物", "电子", "电力设备"}
+        large_cap_weight = sum(distribution.get(ind, 0) for ind in large_cap_industries) / total
+
+        # 价值行业权重
+        value_industries = {
+            "银行",
+            "房地产",
+            "建筑装饰",
+            "公用事业",
+            "交通运输",
+            "煤炭",
+            "石油石化",
+        }
+        value_weight = sum(distribution.get(ind, 0) for ind in value_industries) / total
+
+        market_cap_style = (
+            "大盘" if large_cap_weight > 0.5 else ("小盘" if large_cap_weight < 0.3 else "中盘")
+        )
+        investment_style = (
+            "价值" if value_weight > 0.3 else ("成长" if value_weight < 0.15 else "平衡")
+        )
+
+        return {
+            "market_cap_style": market_cap_style,
+            "investment_style": investment_style,
+            "grid_position": f"{market_cap_style}{investment_style}",
+            "large_cap_weight": round(large_cap_weight * 100, 2),
+            "value_weight": round(value_weight * 100, 2),
+            "industry_distribution": distribution,
+        }
+
+    def get_metrics(self) -> list[str]:
+        """返回支持的指标列表"""
+        return [
+            "industry_distribution",
+            "top_holdings",
+            "concentration_hhi",
+            "concentration_level",
+            "style_analysis",
+        ]
+
+    @staticmethod
+    def _hhi_level(hhi: float) -> str:
+        """根据HHI值返回集中度等级"""
+        if hhi >= 0.25:
+            return "高度集中"
+        elif hhi >= 0.15:
+            return "中度集中"
+        else:
+            return "分散"

fund_cli/analysis/manager.py

@@ -0,0 +1,133 @@
+"""
+基金经理分析引擎
+
+提供基金经理信息查询、业绩统计和稳定性分析功能。
+"""
+
+from typing import Any
+
+from fund_cli.core.analyzer import Analyzer
+
+
+class ManagerAnalyzer(Analyzer):
+    """
+    基金经理分析引擎
+
+    分析基金经理的综合信息，包括：
+    - 经理基本信息查询
+    - 业绩统计（管理基金数、平均收益率等）
+    - 稳定性分析（任职年限、管理规模变化）
+    """
+
+    def analyze(self, data: dict[str, Any], **kwargs: Any) -> dict[str, Any]:
+        """
+        执行综合经理分析
+
+        Args:
+            data: 经理数据字典，需包含 name, fund_code, fund_name 等字段
+
+        Returns:
+            综合分析结果
+        """
+        result = {}
+        result["info"] = self.manager_info(data)
+        result["performance"] = self.performance_stats(data)
+        result["stability"] = self.stability_analysis(data)
+        return result
+
+    def manager_info(self, data: dict[str, Any]) -> dict[str, Any]:
+        """
+        经理信息查询 (FUND-MANAGER-001)
+
+        Args:
+            data: 经理数据字典
+
+        Returns:
+            经理信息字典
+        """
+        return {
+            "name": data.get("name", ""),
+            "fund_code": data.get("fund_code", ""),
+            "fund_name": data.get("fund_name", ""),
+            "company": data.get("company", ""),
+            "start_date": str(data.get("start_date", "")),
+            "tenure_days": data.get("tenure_days", 0),
+        }
+
+    def performance_stats(self, data: dict[str, Any]) -> dict[str, Any]:
+        """
+        经理业绩统计 (FUND-MANAGER-002)
+
+        Args:
+            data: 经理数据字典，可包含 total_return, annual_return, funds 列表
+
+        Returns:
+            业绩统计字典
+        """
+        funds = data.get("funds", [])
+        if funds:
+            returns = [f.get("total_return", 0) for f in funds if f.get("total_return") is not None]
+            avg_return = sum(returns) / len(returns) if returns else 0
+            best_fund = max(funds, key=lambda f: f.get("total_return", float("-inf")), default=None)
+            worst_fund = min(funds, key=lambda f: f.get("total_return", float("inf")), default=None)
+            return {
+                "total_funds": len(funds),
+                "avg_return": round(avg_return, 2),
+                "best_fund": best_fund.get("fund_name", "") if best_fund else "",
+                "best_return": best_fund.get("total_return", 0) if best_fund else 0,
+                "worst_fund": worst_fund.get("fund_name", "") if worst_fund else "",
+                "worst_return": worst_fund.get("total_return", 0) if worst_fund else 0,
+            }
+        else:
+            return {
+                "total_funds": 1,
+                "avg_return": data.get("annual_return", 0),
+                "best_fund": data.get("fund_name", ""),
+                "best_return": data.get("total_return", 0),
+                "worst_fund": data.get("fund_name", ""),
+                "worst_return": data.get("total_return", 0),
+            }
+
+    def stability_analysis(self, data: dict[str, Any]) -> dict[str, Any]:
+        """
+        经理稳定性分析 (FUND-MANAGER-003)
+
+        Args:
+            data: 经理数据字典，可包含 tenure_days, start_date, funds 列表
+
+        Returns:
+            稳定性分析字典
+        """
+        tenure_days = data.get("tenure_days", 0)
+        tenure_years = tenure_days / 365.25 if tenure_days else 0
+
+        # 稳定性评级
+        if tenure_years >= 5:
+            stability_level = "非常稳定"
+            stability_score = 5
+        elif tenure_years >= 3:
+            stability_level = "稳定"
+            stability_score = 4
+        elif tenure_years >= 1:
+            stability_level = "一般"
+            stability_score = 3
+        else:
+            stability_level = "较新"
+            stability_score = 2
+
+        # 多基金管理分析
+        funds = data.get("funds", [])
+        multi_fund = len(funds) > 1
+
+        return {
+            "tenure_days": tenure_days,
+            "tenure_years": round(tenure_years, 1),
+            "stability_level": stability_level,
+            "stability_score": stability_score,
+            "multi_fund_manager": multi_fund,
+            "managed_fund_count": len(funds) if funds else 1,
+        }
+
+    def get_metrics(self) -> list[str]:
+        """返回支持的指标列表"""
+        return ["manager_info", "performance_stats", "stability_analysis"]