detect-palladium-lead-silver-turns

<essential_principles>

<principle name="cross_metal_confirmation"> **Core of Cross-Metal Confirmation**

The hypothesis of "Palladium leads Silver" requires quantifiable verification:

Estimate the optimal lead-lag using cross-correlation
When a turning point appears in silver, check if palladium has a同向 turning point leading or synchronously within the confirmation window
Unconfirmed turning points are candidates for "failed moves"

Lead-Lag = argmax(cross_correlation(pd_ret[t-k:t], ag_ret[t:t+k]))
Confirmed = pd_turn exists within [ag_turn.ts - window, ag_turn.ts + window]

</principle> <principle name="turning_point_detection"> **Three Methods for Turning Point Detection**

Method	Principle	Applicable Scenario
`pivot`	Local extremum within N candlesticks on both sides	Trends with clear structure
`peaks`	scipy find_peaks + prominence	Automated density control
`slope_change`	Trend slope reverses from positive to negative or vice versa	Smooth trend tracking

It is recommended to start with

pivot

, using 3-5 candlesticks on each side, then adjust as needed. </principle>

<principle name="participation_judgment"> **Participation Judgment**

There are multiple ways to measure whether palladium "participates" in silver's trend:

Indicator	Definition	Recommended Threshold
`returns_corr`	Rolling correlation coefficient of returns	> 0.5
`direction_agree`	Proportion of same-direction price movements	> 60%
`vol_expansion`	Synchronized volatility expansion of both	σ_pd / σ_ag > 0.8
`breakout_confirm`	Palladium also breaks out when silver breaks out	Same-direction breakout

When thresholds are not met, silver's movements may be "liquidity noise" rather than trend confirmation. </principle>

<principle name="failure_move_detection"> **Failed Trend Judgment**

Translate "silver movements without palladium participation" into backtestable rules:

Rule	Definition	Consequence
`no_confirm_then_revert`	No confirmation + silver retraces beyond the starting point within N candlesticks	Marked as failed_move
`no_confirm_then_break_fail`	No confirmation + silver breaks out then falls back below the breakout level	False breakout

Historical statistics: Failure rate of unconfirmed events vs success rate of confirmed events. </principle>

</essential_principles>

<objective> Detect cross-metal turning points where "Palladium leads, Silver follows":

Data Acquisition: OHLCV data of Silver and Palladium (yfinance: SI=F, PA=F)
Turning Point Detection: Identify local highs and lows of both (pivot / peaks / slope_change)
Lead-Lag Estimation: Find the optimal lag using cross-correlation
Cross-Metal Confirmation: Check if silver's turning point is confirmed by a same-direction turning point of palladium within the window
Failed Trend Judgment: Check if unconfirmed silver turning points meet the failure rules

Output: Confirmation rate, failure rate, detailed judgment of each event, risk control suggestions. </objective>

<quick_start>

Quickest Way: Detect if recent silver turning points are confirmed by palladium

bash

cd skills/detect-palladium-lead-silver-turns
pip install pandas numpy yfinance scipy statsmodels  # First-time use
python scripts/palladium_lead_silver.py --silver SI=F --palladium PA=F --quick

Output Example:

json

{
  "symbol_pair": {"silver": "SI=F", "palladium": "PA=F"},
  "as_of": "2026-01-14",
  "timeframe": "1h",
  "estimated_pd_leads_by_bars": 6,
  "lead_lag_corr": 0.42,
  "confirmation_rate": 0.71,
  "unconfirmed_failure_rate": 0.64,
  "latest_event": {
    "ts": "2026-01-15T14:00:00Z",
    "turn": "top",
    "confirmed": false,
    "participation_ok": false,
    "failed_move": true
  }
}

Full Analysis:

bash

python scripts/palladium_lead_silver.py --silver SI=F --palladium PA=F --timeframe 1h --lookback 1000 --output result.json

Generate Bloomberg-Style Visualization Chart (Recommended):

bash

pip install matplotlib yfinance  # First-time use
python scripts/plot_bloomberg_style.py --input result.json --output output/palladium_silver_2026-01-26.png

Chart Features:

Bloomberg Professional Color Scheme: Dark background, orange-red silver line, orange-yellow palladium line
Background Band Marking: Green background = Confirmed turning point area, Red background = Unconfirmed turning point area (does not block price lines)
Latest Event Annotation: Prominently marks the confirmation status and price of the latest turning point
Pd/Ag Price Ratio Chart: Shows relative price changes of palladium to silver, including 20-period moving average
Rolling Confirmation Rate: Dynamically displays the validity trend of the confirmation logic
Statistics Panel: Key indicators such as confirmation rate, failure rate, total number of turning points
Market Interpretation: Current status assessment and actionable suggestions

Traditional Three-in-One Chart (Technical Analysis Oriented):

bash

python scripts/plot_palladium_silver.py --silver SI=F --palladium PA=F --output output/

Includes:

Silver/Palladium price overlay with turning point markings
Distribution of confirmed/unconfirmed events
Time series of rolling correlation coefficient
Failed trend statistics

</quick_start>

<intake> What operation do you need to perform?

Quick Detection - Check if the latest silver turning point is confirmed by palladium
Historical Backtest - Backtest the effectiveness of cross-metal confirmation
Continuous Monitoring - Set up alerts to notify when new turning points appear
Parameter Tuning - Find the optimal confirmation window and participation thresholds
Methodology Learning - Understand the theoretical basis of cross-metal lead-lag

Please select or directly provide analysis parameters to start. </intake>

<routing> | Response | Action | |------------------------------------|--------------------------------------------------------| | 1, "quick", "fast", "check" | Execute `python scripts/palladium_lead_silver.py --quick` | | 2, "backtest", "history", "historical" | Read `workflows/backtest.md` and execute | | 3, "monitor", "alert", "watch" | Read `workflows/monitor.md` and execute | | 4, "optimize", "tune", "adjust" | Read the parameter tuning section in `workflows/detect.md` | | 5, "learn", "methodology", "theory" | Read `references/methodology.md` | | Provide parameters (e.g., timeframe, lookback) | Read `workflows/detect.md` and execute with parameters |

After routing, read the corresponding file and execute. </routing>

<directory_structure>

detect-palladium-lead-silver-turns/
├── SKILL.md                           # This file (router)
├── skill.yaml                         # Frontend display metadata
├── manifest.json                      # Skill metadata
├── workflows/
│   ├── detect.md                      # One-time detection workflow
│   ├── backtest.md                    # Historical backtest workflow
│   └── monitor.md                     # Continuous monitoring workflow
├── references/
│   ├── methodology.md                 # Cross-metal lead-lag methodology
│   ├── input-schema.md                # Complete input parameter definition
│   └── data-sources.md                # Data source description
├── templates/
│   ├── output-json.md                 # JSON output structure definition
│   └── output-markdown.md             # Markdown report template
├── scripts/
│   ├── palladium_lead_silver.py       # Main detection script
│   ├── plot_bloomberg_style.py        # Bloomberg-style visualization (Recommended)
│   └── plot_palladium_silver.py       # Traditional three-in-one chart
└── examples/
    └── silver-palladium-2024.json     # Example output

</directory_structure>

<reference_index>

Methodology: references/methodology.md

Cross-metal lead-lag principles
Detailed explanation of three turning point detection methods
Participation and confirmation logic
Market implications of failed trends

Data Sources: references/data-sources.md

Yahoo Finance futures codes
Macro filter data sources
Data frequency and alignment

Input Parameters: references/input-schema.md

Complete parameter definitions
Default values and recommended ranges

</reference_index>

<workflows_index>

Workflow	Purpose	Applicable Scenario
detect.md	One-time detection	Check specific time ranges
backtest.md	Historical backtest	Verify the validity of confirmation logic
monitor.md	Continuous monitoring	Daily tracking or alerts
</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	Purpose
palladium_lead_silver.py	`--silver SI=F --palladium PA=F --quick`	Quick check of current status
palladium_lead_silver.py	`--silver SI=F --palladium PA=F --lookback 1000`	Complete historical analysis
plot_bloomberg_style.py	`--input result.json --output output/chart.png`	Bloomberg-style chart (Recommended)
plot_palladium_silver.py	`--silver SI=F --palladium PA=F --output dir/`	Traditional three-in-one chart
</scripts_index>

<input_schema_summary>

Core Parameters

Parameter	Type	Default Value	Description
silver_symbol	string	(required)	Silver target code
palladium_symbol	string	(required)	Palladium target code
timeframe	string	1h	Analysis time scale
lookback_bars	int	1000	Number of historical candlesticks to look back

Turning Point Detection Parameters

Parameter	Type	Default Value	Description
turn_method	string	pivot	Turning point detection method
pivot_left	int	3	Number of confirmation candlesticks on the left for pivot method
pivot_right	int	3	Number of confirmation candlesticks on the right for pivot method
confirm_window_bars	int	6	Cross-metal confirmation window
lead_lag_max_bars	int	24	Maximum number of lag candlesticks for lead-lag estimation

Participation Parameters

Parameter	Type	Default Value	Description
participation_metric	string	direction_agree	Participation measurement method
participation_threshold	float	0.6	Participation threshold
failure_rule	string	no_confirm_then_revert	Failed trend rule

Complete parameter definition can be found in

references/input-schema.md

</input_schema_summary>

<output_schema_summary>

json

{
  "skill": "detect-palladium-lead-silver-turns",
  "symbol_pair": {"silver": "SI=F", "palladium": "PA=F"},
  "as_of": "2026-01-14",
  "timeframe": "1h",
  "lookback_bars": 1200,
  "summary": {
    "estimated_pd_leads_by_bars": 6,
    "lead_lag_corr": 0.42,
    "confirmation_rate": 0.71,
    "unconfirmed_failure_rate": 0.64,
    "total_ag_turns": 24,
    "confirmed_turns": 17,
    "failed_moves": 5
  },
  "events": [
    {
      "ts": "2026-01-08T10:00:00Z",
      "turn": "bottom",
      "confirmed": true,
      "confirmation_lag_bars": -3,
      "participation_ok": true,
      "failed_move": false
    }
  ],
  "interpretation": {
    "regime_assessment": "...",
    "tactics": ["...", "..."]
  }
}

Complete output structure can be found in

templates/output-json.md

. </output_schema_summary>

<success_criteria> Successful execution should produce:

Optimal lead-lag estimation (in bars) of palladium to silver
Correlation coefficient of lead-lag
Confirmation rate of silver turning points confirmed by palladium
Proportion of failed trends among unconfirmed events
Detailed judgment of each silver turning point (confirmed, participation, failed)
Market interpretation and tactical suggestions
Time series data (optional, for visualization)
Visualization chart PNG (optional, using plot_palladium_silver.py) </success_criteria>

detect-palladium-lead-silver-turns

NPX Install

Tags

SKILL.md Content (Chinese)