メニュー
Track cumulative return of stocks/indices with multi-ticker comparison, index Top N ranking, and visualization. All comparisons use S&P 500 as the fixed benchmark.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/fatfingererr/macro-skills --skill track-equity-cumulative-returnコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
以槓桿反向 ETF（做空）相對槓桿正向 ETF（做多）的交易占比，作為散戶風險偏好代理指標，評估 SPX 後續下行風險。
用勞力士市場指數（WatchCharts Rolex Market Index）作為高 β 的風險偏好／流動性代理，判讀「流動性改善但未到投機狂熱」的狀態
使用公開市場資料量化驗證鉑/白金對巴西股市（EWZ）的長週期傳導/連動關係，輸出雙軸圖、領先落後分析、關聯強度分數與監控訊號。
SKILL.md を表示中
SOC 職業分類に基づく
| name | track-equity-cumulative-return |
| version | 1.1.0 |
| updated | "2026-01-28T00:00:00.000Z" |
| description | Track cumulative return of stocks/indices with multi-ticker comparison, index Top N ranking, and visualization. All comparisons use S&P 500 as the fixed benchmark. |
<essential_principles>
**S&P 500 Fixed Benchmark (Core Methodology)**All cumulative return analyses use S&P 500 (^GSPC) as the fixed benchmark. This is a core methodology decision:
This is hardcoded and cannot be changed.
**Base Date Methodology**For cumulative return calculation, the base date is the last trading day of the previous year:
Cumulative Return = ((Final Price / Base Price) - 1) × 100%
Key methodology:
All tickers are aligned to common trading days with data.
**Four Analysis Scenarios**This skill supports 4 distinct scenarios:
| Scenario |
|---|
| Mode |
|---|
| Description |
|---|
| Example |
|---|
| 1.a | Stock(s), Year Only | Analyze specific tickers for a single full year | NVDA, AMD in 2024 only |
| 1.b | Stock(s), Year to Today | Analyze specific tickers from a year to today | NVDA, AMD from 2022 to today |
| 2.a | Index Top N, Year Only | Rank index components for a single full year | Nasdaq 100 Top N in 2024 only |
| 2.b | Index Top N, Year to Today | Rank index components from a year to today | Nasdaq 100 Top N from 2022 to today |
Use --year-only flag to switch between "Year Only" (a) and "Year to Today" (b) modes.
| Index Code | Name | Components |
|---|---|---|
| nasdaq100 | Nasdaq 100 Index | ~100 |
| sp100 | S&P 100 Index | 100 |
| dow30 | Dow Jones 30 Index | 30 |
| sox | Philadelphia Semiconductor Index | 30 |
Top N analysis fetches all component stocks and ranks by return.
</essential_principles>
Output: Cumulative return time series chart, performance ranking table, JSON data, Markdown report.
<quick_start>
Quick Start: Analyze Stock Cumulative Returns
cd skills/track-equity-cumulative-return/scripts
pip install pandas numpy yfinance matplotlib # First time only
# Scenario 1.a: Stock(s), 2024 Year Only
python cumulative_return_analyzer.py --ticker NVDA AMD --year 2024 --year-only
# Scenario 1.b: Stock(s), 2022 to Today
python cumulative_return_analyzer.py --ticker NVDA AMD GOOGL --year 2022
# Scenario 2.a: Nasdaq 100 Top 10, 2024 Year Only
python index_component_analyzer.py --index nasdaq100 --year 2024 --year-only --top 10
# Scenario 2.b: Nasdaq 100 Top 20, 2022 to Today
python index_component_analyzer.py --index nasdaq100 --year 2022 --top 20
# Visualization (with charts)
python visualize_cumulative.py --ticker NVDA AMD --year 2024 --year-only
python visualize_cumulative.py --mode top20 --index nasdaq100 --year 2022 --top 20
Sample output:
{
"skill": "track-equity-cumulative-return",
"as_of": "2026-01-28",
"mode": "year_to_today",
"parameters": {
"tickers": ["NVDA", "AMD"],
"start_year": 2022,
"year_only": false
},
"benchmark": {
"ticker": "^GSPC",
"name": "S&P 500",
"cumulative_return_pct": 45.2
},
"summary": {
"best_performer": "NVDA",
"best_return": 542.2,
"beat_benchmark_count": 2
}
}
</quick_start>
Scenario Selection:
Provide your analysis parameters or select a scenario.
Key flags:
--year-only: Analyze only the specified year (scenarios a)--year-only: Analyze from year to today (scenarios b)--top N: Select Top N for index analysisAll scripts use Yahoo Finance real data with caching. Benchmark is always S&P 500.
<reference_index>
Reference Documents (references/)
| File | Content |
|---|---|
| methodology.md | Cumulative return calculation methodology |
| data-sources.md | Yahoo Finance data source documentation |
| input-schema.md | Complete input parameter definitions |
| index-components.md | Supported index component lists |
| </reference_index> |
<workflows_index>
| Workflow | Scenario | Use Case |
|---|---|---|
| quick-check.md | 1.a/1.b | Quick check single ticker |
| compare.md | 1.a/1.b | Compare multiple tickers |
| top20.md | 2.a/2.b | Index Top N analysis |
| </workflows_index> |
<templates_index>
| Template | Purpose |
|---|---|
| output-json.md | JSON output structure definition |
| output-markdown.md | Markdown report template |
| </templates_index> |
<scripts_index>
| Script | Command Example | Purpose |
|---|---|---|
| fetch_price_data.py | --ticker NVDA --start 2022-01-01 | Yahoo Finance data fetching |
| cumulative_return_analyzer.py | --ticker NVDA AMD --year 2022 | Cumulative return calculation (1.a/1.b) |
| index_component_analyzer.py | --index nasdaq100 --year 2022 | Index component analysis (2.a/2.b) |
| visualize_cumulative.py | --ticker NVDA AMD --year 2022 | visualization |
| </scripts_index> |
<input_schema_summary>
Required Parameters
| Parameter | Type | Description |
|---|---|---|
| ticker | string | Stock ticker(s) - can be multiple |
| year | int | Start year |
Optional Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
| year-only | flag | false | If set, analyze only the specified year |
| index | string | nasdaq100 | Index type (for Top N mode) |
| top | int | 20 | Top N to select |
| output | string | auto | Output file path |
| mode | string | compare | Mode (compare/top20) |
Note: Benchmark is hardcoded to S&P 500 (^GSPC) and cannot be changed.
See references/input-schema.md for complete parameter definitions.
</input_schema_summary>
Chart Specifications
Charts follow thoughts/shared/guide/bloomberg-style-chart-guide.md:
#1a1a2e (dark blue-black)#2d2d44 (dark gray-purple)#ff6b35 (orange-red), #ffaa00 (orange-yellow)#004E89 (deep blue dashed)#666666 (gray dotted)X-axis format:
Output specs:
<output_schema_summary>
{
"skill": "track-equity-cumulative-return",
"as_of": "2026-01-28",
"mode": "year_to_today",
"parameters": {
"tickers": ["NVDA", "AMD"],
"start_year": 2022,
"year_only": false
},
"period": {
"start_date": "2021-12-31",
"end_date": "2026-01-28",
"years_held": 4.08
},
"benchmark": {
"ticker": "^GSPC",
"name": "S&P 500",
"cumulative_return_pct": 45.2
},
"summary": {
"best_performer": "NVDA",
"best_return": 542.2,
"benchmark_return": 45.2,
"beat_benchmark_count": 2
},
"results": [
{
"ticker": "NVDA",
"name": "NVIDIA (NVDA)",
"cumulative_return_pct": 542.2,
"vs_benchmark": 497.0
}
],
"chart_path": "output/cumulative_return_2026-01-28.png"
}
See templates/output-json.md for complete output structure.
</output_schema_summary>
<success_criteria> Successful execution should produce:
Chart X-axis: Year shown in January, month numbers (2-12) for other months. </success_criteria>
<extended_examples>
Example 1: Single Stock Full Year Analysis (Scenario 1.a)
Analyze NVIDIA's performance in 2024:
cd skills/track-equity-cumulative-return/scripts
python cumulative_return_analyzer.py --ticker NVDA --year 2024 --year-only
Expected output:
==========================================================================================
Cumulative Return Analysis Report
==========================================================================================
Period: 2024 Full Year (2023-12-29 ~ 2024-12-31)
Benchmark: S&P 500
==========================================================================================
Rank Ticker Name Cum. Return vs Bench
----------------------------------------------------------------------
1 NVDA NVIDIA (NVDA) +185.52% +160.97% ✓
----------------------------------------------------------------------
Bench ^GSPC S&P 500 +24.54%
==========================================================================================
Statistics:
- Best performer: NVDA (+185.52%)
- Beat benchmark: 1 / 1
Example 2: Multi-Stock Long-Term Comparison (Scenario 1.b)
Compare FAANG stocks from 2020 to today:
python cumulative_return_analyzer.py --ticker META AAPL AMZN NFLX GOOGL --year 2020
python visualize_cumulative.py --ticker META AAPL AMZN NFLX GOOGL --year 2020
Example 3: Semiconductor Index Top 10 (Scenario 2.a)
Find top 10 semiconductor performers in 2024:
python index_component_analyzer.py --index sox --year 2024 --year-only --top 10
python visualize_cumulative.py --mode top20 --index sox --year 2024 --year-only --top 10
Example 4: Dow 30 Long-Term Analysis (Scenario 2.b)
Analyze Dow 30 components from 2020:
python index_component_analyzer.py --index dow30 --year 2020 --top 30
</extended_examples>
<error_handling>
Input Validation
The skill includes comprehensive input validation:
BRK.B → BRK-B, FB → META)nasdaq100, sp100, dow30, soxNetwork Retry Logic
Yahoo Finance API calls include automatic retry:
Data Quality Checks
</error_handling>
Running Tests
cd skills/track-equity-cumulative-return/scripts/tests
python test_calculations.py
Test Coverage:
Golden Cases
Located in scripts/tests/golden_cases.json:
<data_governance>
Data Sources
| Source | Type | Caching | Notes |
|---|---|---|---|
| Yahoo Finance | Primary | 12-hour cache | Free, public API |
Caching
scripts/cache/python fetch_price_data.py --clear-cacheKnown Limitations
</data_governance>