Anomaly Detection Alerts
Automated detection of suspicious patterns and potential issues in optimization results.
High Severity
1
Medium Severity
0
Low Severity
0
Potential Overfitting Detected
Recommendations
- Review high-severity anomalies before deploying to live trading.
- Consider re-optimizing with different parameters or using Walk-Forward Analysis to validate results.
Overfitting Detection
Analysis of potential overfitting in the optimization results.
Overfitting Detection
Analysis of potential overfitting indicators based on WFA metrics and performance gaps.
High Overfitting Risk Detected
Detailed Analysis
| Indicator | Value | Status | Description | |
|---|---|---|---|---|
| WFA Overfitting Score | 0.0% | Passed | Direct measure of overfitting from Walk-Forward Analysis. Higher values indicate stronger overfitting. | |
| IS vs OOS Return Gap | -119.4% (IS: -2.78%, OOS: 0.54%) | Passed | Large gap between in-sample and out-of-sample returns indicates overfitting. | |
| Walk-Forward Efficiency (WFE) | 19.4% | Failed | Low WFE indicates poor out-of-sample performance relative to in-sample. | |
| Consistency (Profitability Rate) | 100.0% | Passed | Low consistency means only few validation periods were profitable. | |
| Sharpe Ratio Degradation | IS: -0.19, OOS: 0.07 (Δ-0.26) | Passed | Large drop in Sharpe ratio from IS to OOS indicates overfitting. | |
| Win Rate Degradation | -12.0% | Passed | Large drop in win rate from IS to OOS suggests overfitting. |
Recommendations
- Consider simplifying the strategy or reducing parameter complexity
- Increase the sample size for optimization
- Use more robust optimization methods (e.g., Bayesian)
- Test on multiple market regimes
Performance Snapshot
Key metrics derived from the backtest and optimisation results.
Score
Tested at
Submitted by
Result ID
Strategy Parameters
Final parameter set used to generate the optimisation result.
Analysis Coverage
Which advanced analysis pipelines were executed for this result.
Optimisation Setup
Core configuration of the optimisation and the resulting backtest statistics.
Algorithm Configuration
Performance Metrics
Algorithm Performance
Information about the optimization algorithm, search strategy, and performance metrics.
Score Distribution
Distribution of Score values across all optimization results for this strategy and period.
Weighted Score Breakdown
Detailed breakdown of how the optimization score was calculated from individual metrics.
Notes & Flagging
Add notes and flag optimization results for review.
Multi-Objective Optimization
Advanced multi-objective optimization analysis.
Volatility Regimes Performance
Analysis of strategy performance across different volatility regimes.
Market Regime Analysis
Analysis of strategy performance across different market regimes.
Optimization Convergence
Evolution of Score throughout the optimization process. Shows how the algorithm explored the parameter space and improved results.
Degradation Timeline
Evolution of key metrics throughout the optimization process.
Exploration vs Exploitation
Analysis of the optimization strategy balance: Exploration (searching new areas) vs Exploitation (improving found solutions).
Parameter Analysis
Advanced analysis of parameter space coverage, statistical significance, stability, noise, and robustness.
Next Steps Suggestions
Recommendations for further optimization based on current results.
Metric Comparison Table
Compare key metrics across different optimization results.
Local Minima Detection
Identify potential local minima in the optimization landscape.
Side-by-Side Comparison
Compare multiple optimization results side by side.
Parameter Relationships & Constraints
Analyze relationships and constraints between optimization parameters.
Top Results
Best performing optimization results for this strategy and period.
Parameter Sensitivity Analysis
Analyze how each parameter affects the optimization Score. Higher importance indicates stronger correlation with Score.
Parameter Visualization
Interactive visualizations of parameter space including stability zones, 3D plots, contour maps, clustering, correlation analysis, and parallel coordinates.
Pareto Front
Optimal risk/return combinations. Pareto-optimal points represent solutions that cannot be improved in return without increasing risk, or vice versa.
Walk-Forward Analysis
Validation windows for ETHUSDC to test robustness on unseen market data.
IS vs OOS Performance Gap
Comparison of in-sample (optimization) and out-of-sample (validation) performance metrics. Large gaps indicate potential overfitting.
Walk-Forward Efficiency Distribution
Distribution of WFE (Walk-Forward Efficiency) across all validation windows. Higher WFE indicates better robustness.
Consistency Score Breakdown
Detailed breakdown of consistency across different components.
Backtest Breakdown
Detailed backtest metrics, configuration and trade logs.
Export Full Optimization Log
Download complete optimization data and results.
Generate PDF Report
Create a comprehensive PDF report of the optimization results.
Deploy to Demo/Live Trading
Deploy this optimization result to demo or live trading environment.
Best Practices Guide
Learn best practices for optimization and backtesting.
Comments
Add and view comments about this optimization result.