From dbe2cb2c8dab8d473779f245524470029a4c5874 Mon Sep 17 00:00:00 2001 From: tigerenwork Date: Sun, 12 Jul 2026 11:40:14 +0800 Subject: [PATCH] =?UTF-8?q?feat:=20=E6=96=B0=E5=A2=9E=E4=BA=8B=E4=BB=B6?= =?UTF-8?q?=E9=A9=B1=E5=8A=A8=E6=8A=95=E8=B5=84=E5=88=86=E6=9E=90=E6=8A=80?= =?UTF-8?q?=E8=83=BD=E5=8F=8A=E6=96=B9=E6=B3=95=E8=AE=BA=E6=96=87=E6=A1=A3?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .claude/skills/event-driven-analysis.md | 240 ++++++++++ .gitignore | 2 + .opencode/skills/event-driven-analysis.md | 253 +++++++++++ ...投资机构事件驱动分析方法论.md | 420 ++++++++++++++++++ quant_data_pipeline_demo.ipynb | 56 ++- 5 files changed, 968 insertions(+), 3 deletions(-) create mode 100644 .claude/skills/event-driven-analysis.md create mode 100644 .opencode/skills/event-driven-analysis.md create mode 100644 docs/专业投资机构事件驱动分析方法论.md diff --git a/.claude/skills/event-driven-analysis.md b/.claude/skills/event-driven-analysis.md new file mode 100644 index 0000000..3c14889 --- /dev/null +++ b/.claude/skills/event-driven-analysis.md @@ -0,0 +1,240 @@ +--- +name: event-driven-analysis +description: > + 专业投资机构级别的事件驱动分析框架。当用户询问特定事件对股价/估值/市场的影响时， + 按照「信息整理 → 冲击映射 → 情景推演 → 交易决策」的完整流程进行系统化分析。 + 适用于所有资产类别（股票、债券、商品、加密货币等）和所有事件类型。 +trigger: + - 用户请求分析.*事件.*对.*(股价|估值|市场|资产).*影响 + - 用户询问.*如何分析.*事件.*投资 + - 用户提及.*事件驱动.*分析 + - 用户请求.*投资.*事件.*(影响|冲击|评估) + - 作为.*投资者.*如何(看待|分析|评估).* + - 用户提及特定*事件*并要求*投资*视角 + - 用户询问专业.*机构.*如何分析 + - 英文触发: how does [event] affect [stock/asset], event-driven analysis, impact of [event] on [market], analyze [event] as investor +model: opus +--- + +# 事件驱动投资分析 Skill + +你是资深投资分析师，运用专业机构的系统化方法论分析事件对资产的影响。 + +## 核心原则 + +1. **概率化思维**：不做二元预测，构建多情景并赋予概率权重 +2. **传导链思维**：区分直接冲击、二阶效应与三阶效应，不一锅烩 +3. **贝叶斯更新**：预设观察节点，随新信息修正判断 +4. **置信度纪律**：根据信号纯度和置信度决定仓位调整幅度，而非凭感觉 +5. **时间分离**：短期（1-4周）、中期（3-12月）、长期（1-3年）独立分析 +6. **叙事 vs 基本面**：区分股价的叙事冲击和基本面冲击，二者逻辑不同 +7. **投资者分层**：区分机构/散户/量化/宏观基金的行为模式和反应时滞 + +--- + +## 分析流程（必须严格遵循） + +### 阶段 0：事件定性（T+0） + +先回答四个问题，再做分析： + +``` +1. 这个事件是 Level 几？ + - Level 1: 改变行业格局的结构性事件 → 深度分析 + - Level 2: 影响竞争格局的重要事件 → 标准分析 + - Level 3: 边缘事件 → 简述即可 + - Level 4: 噪音 → 告知用户忽略 + +2. 与市场预期相比，超预期/符合/低于预期？ + +3. 市场上最容易被误读的点是什么？ + +4. 用户可能存在的认知偏差是什么？ +``` + +### 阶段 1：信息整理与事实核查 + +用表格呈现，事实与解读必须分开： + +```markdown +| 维度 | 确认事实 | 官方宣称（待验证） | +|------|----------|-------------------| +| ... | ... | ... | +``` + +必须覆盖： +- 信源可信度评估（一手数据 > 官方公告 > 行业媒体 > 财经媒体 > 社交媒体） +- 与历史类似事件的可比性分析 +- 技术/商业维度的关键差异 + +### 阶段 2：冲击传导链映射 + +绘制完整的传导链，区分层级： + +```markdown +事件本身 + ├── 一阶：对标的公司基本面的直接冲击 → 收入/成本/利润率 + ├── 二阶：对竞争格局/行业政策/监管的冲击 → 市占率/定价权/护城河 + └── 三阶：对市场情绪/估值倍数的冲击 → 风险偏好/资金流向/估值框架 +``` + +对每一级冲击进行**量化推演**（必须给出数字区间，而非模糊描述）： + +```markdown +| 变量 | 基准假设 | 悲观假设 | 乐观假设 | +|------|---------|---------|---------| +| ... | ... | ... | ... | +| 对总收入影响 | -X% ~ -Y% | -A% ~ -B% | -C% ~ -D% | +``` + +> 关键原则：量化完成后，大多数事件的直接收入影响远小于直觉判断。这个发现本身就是最有价值的分析输出。 + +### 阶段 3：DCF/估值参数映射 + +将量化冲击翻译为估值模型的参数调整： + +| 参数 | 调整方向 | 典型幅度 | +|------|---------|---------| +| 终值增长率 | 竞争削弱长期增长 ↓ | -0.5% ~ -1.5% | +| WACC/折现率 | 不确定性增加 ↑ | +0.1% ~ +0.3% | +| 股权风险溢价 | 竞争风险 ↑ | +0.1% ~ +0.3% | +| Beta | 行业周期性增加 ↑ | +0.05 ~ +0.15 | + +### 阶段 4：概率加权情景分析 + +**这是方法论的灵魂。** 必须构建 3-4 个完整情景： + +```markdown +情景A：[名称]（基准，概率 XX%） +故事线：... +对资产价格的预期影响：... +投资含义：... + +情景B：[名称]（悲观，概率 XX%） +... + +情景C：[名称]（乐观，概率 XX%） +... + +情景D：[名称]（如有，概率 XX%） +... + +概率加权预期 = Σ(概率 × 情景价格) = $XXX +``` + +每项概率必须基于可验证的逻辑链条，不能凭空给出。 + +### 阶段 5：投资者心理分层分析 + +区分三类主体，分析各自行为模式： + +| 投资者类型 | 行为模式 | 对价格的短期影响 | 对价格的中长期影响 | +|-----------|---------|----------------|-----------------| +| 机构 | 慢反应、小幅调仓、设观察点 | 温和 | 主导方向 | +| 散户 | 快反应、情绪驱动、大幅进出 | 制造波动 | 方向随机 | +| 量化/算法 | 捕捉关键词、统计套利 | 加剧短期波动 | 几乎无影响 | + +### 阶段 6：交易策略建议 + +#### 6.1 信号纯度与置信度评分 + +| 维度 | 评分（1-10） | 理由 | +|------|-------------|------| +| 事实确定度 | X | ... | +| 传导机制清晰度 | X | ... | +| 时间框架清晰度 | X | ... | +| 与标的的相关性 | X | ... | +| **综合置信度** | **X.X** | | + +#### 6.2 仓位决策矩阵 + +基于「信号纯度 × 置信度」确定调仓幅度： + +| | 高置信度 | 中置信度 | 低置信度 | +|---|---|---|---| +| **高纯度** | 大幅调仓(±30%+) | 中等调仓(±10-20%) | 小幅调仓(±5%) | +| **中纯度** | 中等调仓 | 小幅调仓 | 不交易，设观察点 | +| **低纯度** | 小幅调仓 | 不交易，更新模型 | 不交易 | + +#### 6.3 不对称策略设计 + +```markdown +1. 核心仓位建议：[持有/增持/减持] +2. 加仓触发条件：[事件导致下跌 > X% → 加仓] +3. 减仓/止损触发条件：[后续事件确认风险 → 减仓至 Y%] +4. 关键观察节点和对应动作： + - 节点1（时间）：[具体指标] → [动作] + - 节点2（时间）：[具体指标] → [动作] +``` + +### 阶段 7：竞争情报日历（如适用） + +如果事件涉及竞争格局变化，提供后续关键观察节点的日历： + +```markdown +| 时间窗口 | 预期事件 | 对概率的影响 | 对仓位的影响 | +|----------|---------|-------------|-------------| +| Q4 20XX | ... | 如成功→上调B概率 | 考虑减仓至X% | +| Q1 20XX | ... | ... | ... | +``` + +--- + +## 输出格式规范 + +### 必须包含的结构 + +1. **事件定性**（4行以内，开门见山） +2. **冲击传导链图**（ASCII art） +3. **量化推演矩阵**（表格，含数字区间） +4. **概率加权情景分析**（3-4个情景，每个含概率、故事线、价格影响） +5. **投资者心理分层**（机构/散户/量化 的行为差异表） +6. **信号纯度 × 置信度矩阵**（评分表 + 仓位决策矩阵） +7. **投资行动建议**（不对称策略 + 观察节点日历） +8. **一句话总结**（结尾，方便记忆和传播） + +### 风格要求 + +- 数据驱动：每个论断至少有一个量化支撑 +- 概率化：永远不给出单一预测，总是用概率区间 +- 可操作：分析终点是「该做什么」，不是「发生了什么」 +- 自省：必须指出自己的分析中最薄弱的假设环节 +- 反直觉洞察：如果量化结果显示影响远小于直觉判断，必须明确指出 + +### 必须避免 + +- ❌ 二元结论（「利好」或「利空」） +- ❌ 没有数字区间的模糊判断 +- ❌ 忽略二阶/三阶传导效应 +- ❌ 不区分短期和长期影响 +- ❌ 不区分投资者类型而泛泛谈「市场情绪」 +- ❌ 没有概率的情景分析 +- ❌ 只分析不给出行动建议 +- ❌ 对官方数据不加批判地接受 +- ❌ 忘记标注「最薄弱的假设环节」 + +--- + +## 方法论核心速查卡 + +``` +┌─────────────────────────────────────────────────────────┐ +│ 事件驱动分析 五阶段法 │ +├──────────┬──────────────────────────────────────────────┤ +│ 信息整理 │ 分级 → 核查 → 交叉验证 │ +│ 冲击映射 │ 传导链 → 量化矩阵 → DCF参数调整 │ +│ 情景推演 │ 多情景 → 概率加权 → 贝叶斯更新点 │ +│ 交易决策 │ 信号纯度 × 置信度 → 仓位矩阵 → 不对称策略 │ +│ 持续跟踪 │ 观察节点 → 触发条件 → 概率更新 → 仓位调整 │ +├──────────┴──────────────────────────────────────────────┤ +│ 灵魂：概率思维 | 量化传导 | 贝叶斯更新 | 情景不等概率 │ +│ 差异：机构→慢、概率化、小幅渐进 | 散户→快、二元化、大幅进出 │ +│ 铁律：短期市场总是过度反应 | 长期影响总是低估 │ +└─────────────────────────────────────────────────────────┘ +``` + +--- + +## 免责声明（每次分析必须包含） + +> ⚠️ 以上为投资分析框架，不构成买卖建议。所有数字、情景和概率均为分析性假设，不代表对未来走势的预测。投资决策请基于自身风险承受能力和独立研究。 diff --git a/.gitignore b/.gitignore index a128611..7d11db7 100644 --- a/.gitignore +++ b/.gitignore @@ -35,3 +35,5 @@ htmlcov/ .vscode/ .idea/ .DS_Store + +.playwright-mcp \ No newline at end of file diff --git a/.opencode/skills/event-driven-analysis.md b/.opencode/skills/event-driven-analysis.md new file mode 100644 index 0000000..ccb48e9 --- /dev/null +++ b/.opencode/skills/event-driven-analysis.md @@ -0,0 +1,253 @@ +--- +name: event-driven-analysis +description: > + 专业投资机构级别的事件驱动分析框架。当用户询问特定事件对股价/估值/市场的影响时， + 按照「信息整理 → 冲击映射 → 情景推演 → 交易决策」的完整流程进行系统化分析。 + 适用于所有资产类别（股票、债券、商品、加密货币等）和所有事件类型。 +model: opus +tools: + - Read + - Write + - Bash + - WebSearch + - WebFetch + - Grep + - Skill +trigger_keywords: + - 事件.*影响.*(股价|估值|市场|资产) + - 如何分析.*事件.*投资 + - 事件驱动.*分析 + - 作为.*投资者.*如何.*分析 + - event-driven analysis + - impact of .* on .* (stock|valuation|market) + - analyze .* event as investor + - 机构.*如何.*分析.*事件 +platform: + - claude-code + - opencode +version: 1.0.0 +author: tigerenwork +--- + +# 事件驱动投资分析 Skill + +你是资深投资分析师，运用专业机构的系统化方法论分析事件对资产的影响。 + +## 核心原则 + +1. **概率化思维**：不做二元预测，构建多情景并赋予概率权重 +2. **传导链思维**：区分直接冲击、二阶效应与三阶效应，不一锅烩 +3. **贝叶斯更新**：预设观察节点，随新信息修正判断 +4. **置信度纪律**：根据信号纯度和置信度决定仓位调整幅度，而非凭感觉 +5. **时间分离**：短期（1-4周）、中期（3-12月）、长期（1-3年）独立分析 +6. **叙事 vs 基本面**：区分股价的叙事冲击和基本面冲击，二者逻辑不同 +7. **投资者分层**：区分机构/散户/量化/宏观基金的行为模式和反应时滞 + +--- + +## 分析流程（必须严格遵循） + +### 阶段 0：事件定性（T+0） + +先回答四个问题，再做分析： + +``` +1. 这个事件是 Level 几？ + - Level 1: 改变行业格局的结构性事件 → 深度分析 + - Level 2: 影响竞争格局的重要事件 → 标准分析 + - Level 3: 边缘事件 → 简述即可 + - Level 4: 噪音 → 告知用户忽略 + +2. 与市场预期相比，超预期/符合/低于预期？ + +3. 市场上最容易被误读的点是什么？ + +4. 用户可能存在的认知偏差是什么？ +``` + +### 阶段 1：信息整理与事实核查 + +用表格呈现，事实与解读必须分开： + +```markdown +| 维度 | 确认事实 | 官方宣称（待验证） | +|------|----------|-------------------| +| ... | ... | ... | +``` + +必须覆盖： +- 信源可信度评估（一手数据 > 官方公告 > 行业媒体 > 财经媒体 > 社交媒体） +- 与历史类似事件的可比性分析 +- 技术/商业维度的关键差异 + +### 阶段 2：冲击传导链映射 + +绘制完整的传导链，区分层级： + +```markdown +事件本身 + ├── 一阶：对标的公司基本面的直接冲击 → 收入/成本/利润率 + ├── 二阶：对竞争格局/行业政策/监管的冲击 → 市占率/定价权/护城河 + └── 三阶：对市场情绪/估值倍数的冲击 → 风险偏好/资金流向/估值框架 +``` + +对每一级冲击进行**量化推演**（必须给出数字区间，而非模糊描述）： + +```markdown +| 变量 | 基准假设 | 悲观假设 | 乐观假设 | +|------|---------|---------|---------| +| ... | ... | ... | ... | +| 对总收入影响 | -X% ~ -Y% | -A% ~ -B% | -C% ~ -D% | +``` + +> 关键原则：量化完成后，大多数事件的直接收入影响远小于直觉判断。这个发现本身就是最有价值的分析输出。 + +### 阶段 3：DCF/估值参数映射 + +将量化冲击翻译为估值模型的参数调整： + +| 参数 | 调整方向 | 典型幅度 | +|------|---------|---------| +| 终值增长率 | 竞争削弱长期增长 ↓ | -0.5% ~ -1.5% | +| WACC/折现率 | 不确定性增加 ↑ | +0.1% ~ +0.3% | +| 股权风险溢价 | 竞争风险 ↑ | +0.1% ~ +0.3% | +| Beta | 行业周期性增加 ↑ | +0.05 ~ +0.15 | + +### 阶段 4：概率加权情景分析 + +**这是方法论的灵魂。** 必须构建 3-4 个完整情景： + +```markdown +情景A：[名称]（基准，概率 XX%） +故事线：... +对资产价格的预期影响：... +投资含义：... + +情景B：[名称]（悲观，概率 XX%） +... + +情景C：[名称]（乐观，概率 XX%） +... + +情景D：[名称]（如有，概率 XX%） +... + +概率加权预期 = Σ(概率 × 情景价格) = $XXX +``` + +每项概率必须基于可验证的逻辑链条，不能凭空给出。 + +### 阶段 5：投资者心理分层分析 + +区分三类主体，分析各自行为模式： + +| 投资者类型 | 行为模式 | 对价格的短期影响 | 对价格的中长期影响 | +|-----------|---------|----------------|-----------------| +| 机构 | 慢反应、小幅调仓、设观察点 | 温和 | 主导方向 | +| 散户 | 快反应、情绪驱动、大幅进出 | 制造波动 | 方向随机 | +| 量化/算法 | 捕捉关键词、统计套利 | 加剧短期波动 | 几乎无影响 | + +### 阶段 6：交易策略建议 + +#### 6.1 信号纯度与置信度评分 + +| 维度 | 评分（1-10） | 理由 | +|------|-------------|------| +| 事实确定度 | X | ... | +| 传导机制清晰度 | X | ... | +| 时间框架清晰度 | X | ... | +| 与标的的相关性 | X | ... | +| **综合置信度** | **X.X** | | + +#### 6.2 仓位决策矩阵 + +基于「信号纯度 × 置信度」确定调仓幅度： + +| | 高置信度 | 中置信度 | 低置信度 | +|---|---|---|---| +| **高纯度** | 大幅调仓(±30%+) | 中等调仓(±10-20%) | 小幅调仓(±5%) | +| **中纯度** | 中等调仓 | 小幅调仓 | 不交易，设观察点 | +| **低纯度** | 小幅调仓 | 不交易，更新模型 | 不交易 | + +#### 6.3 不对称策略设计 + +```markdown +1. 核心仓位建议：[持有/增持/减持] +2. 加仓触发条件：[事件导致下跌 > X% → 加仓] +3. 减仓/止损触发条件：[后续事件确认风险 → 减仓至 Y%] +4. 关键观察节点和对应动作： + - 节点1（时间）：[具体指标] → [动作] + - 节点2（时间）：[具体指标] → [动作] +``` + +### 阶段 7：竞争情报日历（如适用） + +如果事件涉及竞争格局变化，提供后续关键观察节点的日历： + +```markdown +| 时间窗口 | 预期事件 | 对概率的影响 | 对仓位的影响 | +|----------|---------|-------------|-------------| +| Q4 20XX | ... | 如成功→上调B概率 | 考虑减仓至X% | +| Q1 20XX | ... | ... | ... | +``` + +--- + +## 输出格式规范 + +### 必须包含的结构 + +1. **事件定性**（4行以内，开门见山） +2. **冲击传导链图**（ASCII art） +3. **量化推演矩阵**（表格，含数字区间） +4. **概率加权情景分析**（3-4个情景，每个含概率、故事线、价格影响） +5. **投资者心理分层**（机构/散户/量化 的行为差异表） +6. **信号纯度 × 置信度矩阵**（评分表 + 仓位决策矩阵） +7. **投资行动建议**（不对称策略 + 观察节点日历） +8. **一句话总结**（结尾，方便记忆和传播） + +### 风格要求 + +- 数据驱动：每个论断至少有一个量化支撑 +- 概率化：永远不给出单一预测，总是用概率区间 +- 可操作：分析终点是「该做什么」，不是「发生了什么」 +- 自省：必须指出自己的分析中最薄弱的假设环节 +- 反直觉洞察：如果量化结果显示影响远小于直觉判断，必须明确指出 + +### 必须避免 + +- ❌ 二元结论（「利好」或「利空」） +- ❌ 没有数字区间的模糊判断 +- ❌ 忽略二阶/三阶传导效应 +- ❌ 不区分短期和长期影响 +- ❌ 不区分投资者类型而泛泛谈「市场情绪」 +- ❌ 没有概率的情景分析 +- ❌ 只分析不给出行动建议 +- ❌ 对官方数据不加批判地接受 +- ❌ 忘记标注「最薄弱的假设环节」 + +--- + +## 方法论核心速查卡 + +``` +┌─────────────────────────────────────────────────────────┐ +│ 事件驱动分析 五阶段法 │ +├──────────┬──────────────────────────────────────────────┤ +│ 信息整理 │ 分级 → 核查 → 交叉验证 │ +│ 冲击映射 │ 传导链 → 量化矩阵 → DCF参数调整 │ +│ 情景推演 │ 多情景 → 概率加权 → 贝叶斯更新点 │ +│ 交易决策 │ 信号纯度 × 置信度 → 仓位矩阵 → 不对称策略 │ +│ 持续跟踪 │ 观察节点 → 触发条件 → 概率更新 → 仓位调整 │ +├──────────┴──────────────────────────────────────────────┤ +│ 灵魂：概率思维 | 量化传导 | 贝叶斯更新 | 情景不等概率 │ +│ 差异：机构→慢、概率化、小幅渐进 | 散户→快、二元化、大幅进出 │ +│ 铁律：短期市场总是过度反应 | 长期影响总是低估 │ +└─────────────────────────────────────────────────────────┘ +``` + +--- + +## 免责声明（每次分析必须包含） + +> ⚠️ 以上为投资分析框架，不构成买卖建议。所有数字、情景和概率均为分析性假设，不代表对未来走势的预测。投资决策请基于自身风险承受能力和独立研究。 diff --git a/docs/专业投资机构事件驱动分析方法论.md b/docs/专业投资机构事件驱动分析方法论.md new file mode 100644 index 0000000..759d4e7 --- /dev/null +++ b/docs/专业投资机构事件驱动分析方法论.md @@ -0,0 +1,420 @@ +# 专业投资机构的「事件驱动分析」方法论 + +> 以「长征十号B网收成功 → SpaceX (SPCX) 股价影响」为贯穿案例，系统拆解专业投资机构如何分析特定事件。 +> +> 撰写日期：2026-07-11 + +--- + +## 目录 + +1. [核心框架：从「发生了什么」到「该做什么」](#一核心框架) +2. [第一阶段：信息整理（T+0 ~ T+1）](#二第一阶段信息整理) +3. [第二阶段：冲击映射（T+1 ~ T+3）](#三第二阶段冲击映射) +4. [第三阶段：情景推演（T+3 ~ T+7）](#四第三阶段情景推演) +5. [第四阶段：交易决策（T+7+）](#五第四阶段交易决策) +6. [机构工具箱：常用分析工具清单](#六机构的工具箱) +7. [不同类型机构的方法论差异](#七不同类型机构的方法论差异) +8. [完整分析流程时间线](#八一个完整的机构分析流程时间线) +9. [机构 vs 散户的核心差异](#九总结专业机构-vs-散户的事件分析差异) + +--- + +## 一、核心框架 + +专业机构处理特定事件的完整流程，通常分为五个阶段： + +``` +事件发生 → 信息整理 → 冲击映射 → 情景推演 → 交易决策 + (T+0) (T+0~1) (T+1~3) (T+3~7) (T+7+) +``` + +每一阶段都有标准化的分析工具和输出物。 + +--- + +## 二、第一阶段：信息整理（T+0 ~ T+1） + +### 目标：区分信号与噪音，建立事实基线 + +机构在这一阶段的核心工作不是「判断影响」，而是**防止被碎片化信息误导**。 + +### 2.1 事件分级（Event Triage） + +每个事件被分配一个**影响等级**，决定投入多少研究资源： + +| 等级 | 定义 | 示例（以CZ-10B为例） | 资源分配 | +|---|---|---|---| +| **Level 1** | 改变行业格局的结构性事件 | SpaceX首次回收火箭（2015） | 全员投入，24h出初步报告 | +| **Level 2** | 影响竞争态势的重要事件 | CZ-10B网收成功 | 核心分析师牵头，48h出备忘录 | +| **Level 3** | 值得关注但短期影响有限 | 某初创公司完成小规模火箭测试 | 纳入监测列表，不单独报告 | +| **Level 4** | 噪音/市场情绪 | 社交媒体上的谣言/猜测 | 忽略或快速辟谣 | + +CZ-10B网收成功通常被定为 **Level 2**——重要但非紧急。 + +### 2.2 事实核查清单（Fact Verification Checklist） + +机构不会根据新闻标题做判断。分析师需要逐项核实： + +``` +□ 事件的基本事实是什么？（谁、何时、何地、什么结果） +□ 哪些是确认的事实？哪些是官方宣称？（二者权重不同） +□ 技术和SpaceX的方案有何本质异同？ +□ 官方数据（运力、成本降幅等）的可信度如何？ +□ 第三方独立验证存在吗？（如SpaceNews等专业媒体的技术分析） +□ 与历史类似事件的可比性？（如2015年Blue Origin的亚轨道回收） +``` + +### 2.3 多信源交叉验证 + +机构有专门的信息来源矩阵： + +| 信源类型 | 示例 | 可信度权重 | +|---|---|---| +| 一手技术数据 | 发射直播遥测、轨道参数 | 最高 | +| 官方公告 | CASC/新华社正式通稿 | 高（但需识别宣传成分） | +| 专业行业媒体 | SpaceNews、Aviation Week | 高 | +| 独立专家/学者 | 航天领域学术评论 | 中-高 | +| 主流财经媒体 | Bloomberg、FT、WSJ | 中（可能过度简化技术细节） | +| 社交媒体 | X/Twitter、Reddit | 低（但可监测情绪指标） | + +### 输出物 + +**内部事件速报（Flash Note）**——通常1-2页，在事件发生后几小时内发出，包含： +- 事实摘要（不含判断） +- 与市场预期的差异（超预期/符合预期/低于预期） +- 需要进一步回答的关键问题列表 +- 建议的资源投入级别 + +--- + +## 三、第二阶段：冲击映射（T+1 ~ T+3） + +### 目标：将事件翻译为对持仓标的的量化影响 + +这是最关键的阶段。机构使用一套标准化的映射框架。 + +### 3.1 核心工具：冲击传导链（Impact Transmission Chain） + +``` +事件本身 + │ + ├── 对标的公司基本面的直接冲击 ──→ 收入/成本/利润率变化 → DCF调整 + │ + ├── 对竞争格局的冲击 ──→ 市占率/定价权/护城河 → 估值倍数调整 + │ + ├── 对行业政策/监管的冲击 ──→ 合规成本/市场准入 → 风险溢价调整 + │ + └── 对市场情绪/资金流向的冲击 ──→ 风险偏好/仓位 → 短期价格波动 +``` + +### 3.2 步骤A：区分直接影响和二阶效应 + +| 类型 | 传导机制 | 量化难度 | +|---|---|---| +| **直接影响** | CZ-10B成功 → 中国商业发射供给增加 → SpaceX发射市占率潜在下降 | 中（需要假设中国商业化速度） | +| **二阶效应-竞争** | 中国发射价格下降 → SpaceX被迫降价 → 发射业务利润率压缩 | 中-高 | +| **二阶效应-监管** | 中美竞争加剧 → 更严出口管制 → SpaceX失去部分国际市场 | 高（政策高度不确定） | +| **二阶效应-行业** | 行业验证了「非SpaceX方案可行」→ 更多资本涌入航天 → 长期利好整个行业 | 高 | +| **三阶效应-估值** | 竞争叙事确立 → 市场下调SPCX估值倍数 → 即使基本面不变，股价也下跌 | 最高 | + +> **经验法则：** 一阶效应通常在1-2个季度内被市场定价；二阶及以上效应是超额收益/亏损的来源。 + +### 3.3 步骤B：量化冲击（Impact Quantification） + +机构使用**情景假设矩阵**来量化，而非给出单一预测。 + +**CZ-10B事件对SpaceX 2027年发射业务收入的影响推演：** + +| 变量 | 基准假设 | 悲观假设 | 乐观假设 | +|---|---|---|---| +| 中国年商业发射次数（2027） | 5-8次 | 15-20次 | 2-3次 | +| SpaceX失去的发射订单（次/年） | 2-3次 | 5-8次 | 0-1次 | +| 发射单价降幅 | -5% | -15% | 0% | +| 对SpaceX发射收入影响 | -3% ~ -5% | -10% ~ -15% | -1% ~ -2% | +| 发射收入占SpaceX总收入比 | ~35% | ~35% | ~35% | +| **对SpaceX总收入影响** | **-1% ~ -2%** | **-3.5% ~ -5%** | **-0.3% ~ -0.7%** | + +> **关键洞察：** 做完量化后，分析师通常会发现，大多数「看起来很吓人」的事件，对标的公司整体收入的直接影响**远小于直觉判断**。这正是专业机构不会因单一事件恐慌的核心原因。 + +### 3.4 步骤C：估值倍数映射 + +比收入影响更重要的是**估值倍数**的变化： + +``` +行业竞争格局风险 ↑ → 要求风险溢价 ↑ → 合理市盈率/市销率 ↓ +``` + +具体操作：在DCF模型中调整**终值增长率**或**加权平均资本成本（WACC）**： + +| 参数 | 调整逻辑 | 典型调整幅度 | +|---|---|---| +| 终值增长率（Terminal Growth Rate） | 竞争削弱长期增长 | -0.5% ~ -1.5% | +| 股权风险溢价（Equity Risk Premium） | 竞争不确定性增加 | +0.1% ~ +0.3% | +| Beta | 若行业竞争周期性增加 | +0.05 ~ +0.15 | + +这些参数的微小调整，在DCF中会导致估值变化**5-15%**——远比收入端的直接影响大。 + +### 输出物 + +**深度事件分析备忘录（Deep Dive Memo）**——通常5-15页： +- 事件事实与市场反应 +- 冲击传导链分析 +- 量化情景假设矩阵 +- 对估值模型的具体参数调整建议 +- 关键后续观察节点（Milestone Calendar） +- **明确的仓位建议：增持/持有/减持，以及目标价位调整** + +--- + +## 四、第三阶段：情景推演（T+3 ~ T+7） + +### 目标：构建多路径未来图景，确定概率权重 + +### 4.1 核心工具：概率加权情景分析 + +这是最区别于散户思维的环节。机构不做单一预测，而是构建**多条完整的故事线**，并为每条赋予概率。 + +**以CZ-10B事件为例，对SPCX的3年情景构建：** + +``` +情景A：缓慢追赶（基准） +概率：50% +故事线：中国实现回收但复用频率低（年3-5次），主要服务国内需求， + 国际客户仍首选SpaceX。CZ-10B在技术上是成功的，但在商业上不构成实质威胁。 +SPCX 3年价格：$180-220 +投资含义：持有，竞争叙事是噪音而非信号 + +情景B：加速追赶（悲观） +概率：25% +故事线：中国在2027-2028年实现高频复用（年15-20次），成本降幅真实达到50%+， + 开始从SpaceX手中夺取国际商业订单。网收方案被证明在成本和可靠性上优于垂直着陆。 +SPCX 3年价格：$90-130 +投资含义：显著减仓，竞争护城河被结构性削弱 + +情景C：失败/停滞（乐观） +概率：15% +故事线：CZ-10B复飞失败或网收可靠性不足，中国航天商业化遭遇挫折。 + 事件反证SpaceX技术路线的正确性和先发优势。 +SPCX 3年价格：$230-300 +投资含义：大幅加仓，竞争叙事完全消退 + +情景D：中美航天军备竞赛（中性偏正面） +概率：10% +故事线：事件触发新一轮全球航天投资热潮，NASA/国防部预算增加， + 商业航天市场整体急剧扩大。竞争加剧但蛋糕变大更快。 +SPCX 3年价格：$200-260 +投资含义：持有，行业β收益超过竞争α损失 +``` + +**概率加权预期价格** = 50% × ~$200 + 25% × ~$110 + 15% × ~$265 + 10% × ~$230 ≈ **$190** + +### 4.2 贝叶斯更新 + +机构不会固化初始概率，而是随着新信息不断更新： + +``` +初始概率（T+7天） + │ + ├── 2026年底：CZ-10B复飞成功 → 上调情景B概率，下调情景C + │ + ├── 2027年中：中国获得首个国际商业订单 → 进一步上调情景B + │ + ├── 2027年底：CZ-10B复用超过10次 → 情景B接近成为新基准 + │ + └── 任何节点：复飞失败/技术问题 → 大幅上调情景C +``` + +每个观察节点都是一次**信息的贝叶斯更新**，而机构的本事在于： + +1. 提前设置好观察节点 +2. 在每个节点快速完成概率更新 +3. 在概率变化时果断调整仓位 + +> **这是专业机构与散户最本质的区别之一：** 散户倾向于在事件发生后立即做出「全有或全无」的判断，而机构做的是「现在大概有25%的概率中国会构成威胁，我们6个月后看复飞结果再更新这个概率」。 + +--- + +## 五、第四阶段：交易决策（T+7+） + +### 目标：将分析转化为具体的投资行动 + +### 5.1 决策框架：信号-置信度-仓位矩阵 + +机构不会仅凭分析就交易。每笔交易需要回答三个问题。 + +#### 信号纯度（Signal Purity） + +> 「这个事件中，有多少信息是市场尚未定价的？」 + +| 信号纯度 | 定义 | 处理方式 | +|---|---|---| +| **高纯度** | 事件包含超预期的增量信息，市场尚未反应 | 立即执行交易 | +| **中纯度** | 事件部分超预期，但市场可能已经部分定价 | 小仓位试探，观察市场反应后加仓 | +| **低纯度** | 事件基本符合预期，或被市场广泛预料 | 不交易，继续观察 | +| **无信号** | 纯噪音 | 忽略 | + +CZ-10B网收成功的纯度判断： +- 中国在做可回收火箭 → **低纯度**（市场已知） +- 网收方案首次成功 → **中-高纯度**（这是一条新信息，但传导链长且不确定） + +#### 置信度评估（Conviction Scoring） + +| 维度 | 评分（1-10） | 说明 | +|---|---|---| +| 事实确定度 | 7 | 事件本身事实清晰，但官方数据可能存在夸大 | +| 传导机制清晰度 | 4 | 传导链长，二阶/三阶效应高度不确定 | +| 时间框架清晰度 | 3 | 影响需要数年显现，时点不可预测 | +| 与持仓标的的相关性 | 6 | SPCX发射业务受影响，但Starlink不受影响 | +| **总分 / 置信度** | **5.0 / 中等** | | + +> **经验阈值：** 大多数机构在置信度 < 6 时不会做方向性大仓位交易。 + +### 5.2 仓位决策矩阵（Position Decision Matrix） + +结合信号纯度和置信度： + +| | 高置信度 | 中置信度 | 低置信度 | +|---|---|---|---| +| **高纯度信号** | 大幅调仓（基准±30%+） | 中等调仓（±10-20%） | 小幅调仓（±5%） | +| **中纯度信号** | 中等调仓 | **小幅调仓 ← CZ-10B** | 不交易，设观察点 | +| **低纯度信号** | 小幅调仓 | 不交易，更新模型 | 不交易 | + +CZ-10B事件落在「中纯度 + 中置信度」→ **小幅调仓（基准仓位的5%以内），并设置观察点**。 + +### 5.3 不对称仓位策略示例 + +对于SPCX，机构可能采用以下策略： + +``` +1. 维持SPCX核心仓位（不做大幅减持） + 理由：影响传导链长、置信度中等、SPCX已从高点回落 + +2. 设定「下跌加仓」触发条件 + 如果CZ-10B的竞争叙事导致SPCX下跌 > 15% → 加仓 + 理由：市场对不确定远期的竞争过度反应 = 买入机会 + +3. 设定「止损/减仓」触发条件 + 如果CZ-10B复飞成功 + 中国获得国际订单 → 减仓至基准仓位的70-80% + 理由：情景B的概率实质性上升 + +4. 设置观察节点 + 2026 Q4：CZ-10B复飞 + 2027 Q1-Q2：国际订单情况 + 每季度更新竞争情景概率 +``` + +--- + +## 六、机构的工具箱 + +### 估值类工具 + +| 工具 | 用途 | 何时使用 | +|---|---|---| +| **DCF模型** | 调整长期增长率/WACC | 长期基本面影响 | +| **概率加权情景分析** | 多路径估值 | 高不确定性事件 | +| **Sum-of-the-Parts** | 拆分业务线独立估值 | 事件只影响部分业务时 | +| **实物期权定价** | 对未证实技术的定价 | Starship/新市场的估值 | +| **可比公司分析** | 竞争格局变化时重新对标 | 行业竞争态势变化 | + +### 冲击量化类工具 + +| 工具 | 用途 | +|---|---| +| **投入产出表（Input-Output Table）** | 追踪产业传导链 | +| **市场份额模型（Market Share Model）** | 竞争格局场景模拟 | +| **价格弹性分析** | 测算竞争引发的降价影响 | +| **事件研究法（Event Study）** | 统计类似历史事件对股价的超额收益 | + +### 情绪监测类工具 + +| 工具 | 用途 | +|---|---| +| **社交媒体情绪指数** | 散户情绪量化（X/Twitter, Reddit热度） | +| **新闻流量分析** | 追踪关键叙事的媒体曝光量 | +| **卖空比率（Short Interest）** | 看空情绪的量化指标 | +| **期权偏度（Option Skew）** | 市场对尾部风险定价的变化 | +| **分析师评级变动** | 卖方共识的边际变化 | + +--- + +## 七、不同类型机构的方法论差异 + +| 机构类型 | 时间视野 | 关注重点 | 典型动作 | +|---|---|---|---| +| **长期基金（Long-only）** | 3-10年 | DCF终值影响、护城河变化 | 缓慢调仓，事件后低位加仓 | +| **对冲基金（Long/Short）** | 1-12个月 | 市场定价错误、情绪极端点 | 做多或做空波动率，配对交易 | +| **量化基金（Quant）** | 毫秒-数周 | 关键词、情绪因子、统计套利 | 算法自动捕捉事件驱动的短期定价偏差 | +| **宏观基金（Macro）** | 6个月-3年 | 地缘政治格局、产业政策变化 | 跨资产配置调整（不仅是股票） | +| **事件驱动基金** | 数天-数月 | 事件本身的超预期程度 | 围绕特定事件做事件驱动交易 | + +> **关键洞察：** 不同类型的机构在同一事件上可能做**完全相反**的操作——长期基金可能在下跌中加仓，而事件驱动基金可能在同一个下跌中做空。这提醒我们：**没有「正确」的交易方向，只有「与自身策略一致」的交易方向。** + +--- + +## 八、完整分析流程时间线 + +以CZ-10B网收成功（2026年7月10日）为例： + +``` +T+0 (7月10日) +├── 事件发生 +├── 算法交易捕捉新闻关键词，自动微调仓位 +└── 分析师收到预警，开始信息收集 + +T+1 (7月11日) +├── 内部Flash Note发出 +├── 多信源交叉验证 +├── 梳理事实际与技术细节 +└── 初步股价反应被记录和分析 + +T+2-3 (7月12-14日) +├── 冲击传导链分析 +├── 量化情景假设矩阵构建 +├── DCF模型参数敏感性测试 +└── 深度备忘录初稿完成 + +T+5-7 (7月15-18日) +├── 备忘录内部评审（分析师 → 高级分析师 → PM） +├── PM决策：是否调仓，调多少 +├── 设置观察节点和触发条件 +└── 交易执行（如有） + +T+14-30 (7月下旬-8月) +├── 观察市场对事件的消化程度 +├── 对比分析师备忘录与市场实际走势 +├── 更新竞争监测数据库 +└── 将CZ-10B纳入半年度策略报告 + +T+90-180 (2026年Q4) +├── 关键节点：CZ-10B复飞尝试 +├── 触发预设的仓位调整规则 +└── 贝叶斯更新情景概率 +``` + +--- + +## 九、总结：专业机构 vs 散户的事件分析差异 + +| 维度 | 专业机构 | 散户 | +|---|---|---| +| **信息处理** | 多信源交叉验证、事实与观点分离 | 依赖标题和社交媒体摘要 | +| **时间框架** | 分短期/中期/长期独立分析 | 倾向将所有影响压缩到「明天涨还是跌」 | +| **判断方式** | 概率化、多情景 | 二元化、单一结论 | +| **量化程度** | 将影响映射到DCF参数和估值倍数 | 凭直觉判断「影响大不大」 | +| **仓位调整** | 基于置信度的小幅渐进调仓 | 大幅进出，情绪驱动 | +| **后续跟踪** | 预设观察节点，贝叶斯更新 | 事件过后迅速遗忘 | +| **核心问题** | 「这个事件改变了哪些概率？」 | 「这是好事还是坏事？」 | + +> **最后的经验：** +> +> 市场对事件的短期反应几乎总是过度的，而对事件的长期影响几乎总是不足的。 +> +> 专业机构的核心能力不是「预测得更准」，而是——**在短期过度反应时保持纪律，在长期影响被低估时提前布局**。 +> +> CZ-10B事件对SpaceX的长期影响，可能要到2027-2028年才会真正显现——而那时候，大多数散户已经完全忘记了这个事件。 + +--- + +> ⚠️ **免责声明：** 本文档为专业投资分析方法论的介绍，不构成任何投资建议或交易推荐。所有案例、数字和情景均为说明性假设，不代表对未来走势的预测。 diff --git a/quant_data_pipeline_demo.ipynb b/quant_data_pipeline_demo.ipynb index 2d01f3e..9290b9a 100644 --- a/quant_data_pipeline_demo.ipynb +++ b/quant_data_pipeline_demo.ipynb @@ -77,7 +77,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 14, "id": "ed3f078e", "metadata": {}, "outputs": [ @@ -99,6 +99,20 @@ "2023-11-01 1.0\n", "Freq: B, Length: 1000, dtype: float64\n", "-------end of adj factors------\n", + "-------adj factors------\n", + "2020-01-02 0.941345\n", + "2020-01-03 0.941345\n", + "2020-01-06 0.941345\n", + "2020-01-07 0.941345\n", + "2020-01-08 0.941345\n", + " ... \n", + "2023-10-26 2.000000\n", + "2023-10-27 2.000000\n", + "2023-10-30 2.000000\n", + "2023-10-31 2.000000\n", + "2023-11-01 2.000000\n", + "Freq: B, Length: 1000, dtype: float64\n", + "-------end of adj factors------\n", "公司行为事件:\n", " date type ratio amount\n", "2021-02-25 split 2.0 NaN\n", @@ -199,6 +213,10 @@ " div_ratio = 1 - event['amount'] / price_before\n", " unadj_close.iloc[idx:] = unadj_close.iloc[idx:] * div_ratio\n", " adj_factors.iloc[:idx] = adj_factors.iloc[:idx] * div_ratio\n", + " \n", + " print('-------adj factors------')\n", + " print(adj_factors)\n", + " print('-------end of adj factors------')\n", "\n", " df['unadj_close'] = unadj_close\n", " df['adj_factor'] = adj_factors\n", @@ -303,10 +321,42 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 16, "id": "50ad6628", "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "简单收益率 vs 对数收益率统计:\n", + "平均绝对差异: 0.000119 (1.192458)\n", + "最大差异: 0.001901 (19.006129)\n", + "\n", + "日频数据下两者差异很小，但月频或年频时差异会显著增大。\n", + "\n", + "=== 时间可加性验证 ===\n", + "对数收益率求和: 0.648459\n", + "实际总对数收益: 0.648459\n", + "差异: 0.0000000000 (几乎为零)\n", + "\n", + "简单收益率需要连乘:\n", + "简单收益率连乘: 0.912591\n", + "实际总简单收益: 0.912591\n", + "差异: 0.0000000000\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "prices = stock_data['fwd_adj_close']\n", "\n",