ALEEP Config

Configure ML model features, parameters, exit strategies, and run custom analysis

Enterprise Feature

This feature is available on the Enterprise plan. The documentation below describes its full capabilities so you can evaluate whether it fits your workflow. Contact us to learn more about upgrading your access.

Overview

ALEEP Config is the advanced configuration interface for the ALEEP Machine Learning engine. It allows you to customize the ML model's feature selection, tune parameters, configure exit strategies, set entry filters, and run custom analysis on any stock symbol. This is the control center for fine-tuning the ML signal generation pipeline.

Quick Start

●Enter a stock symbol in the search bar
●Select the technical features (indicators) you want the model to use
●Configure ML parameters (K neighbors, prediction days, lookback bars)
●Click "Run Analysis" to generate signals and backtest results
●Review the backtest statistics and signal history

Tip

Start with default settings to establish a baseline. Then adjust one parameter at a time to understand how each change affects signal quality and backtest results.

Feature Selection

Features are the technical indicators the ML model uses as inputs for pattern matching. Each feature captures a different aspect of price action, momentum, or volatility.

Feature	Full Name	What It Measures
RSI	Relative Strength Index	Momentum oscillator measuring speed of price changes (0-100)
WT	WaveTrend	Cross-based momentum indicator with overbought/oversold levels
MFI	Money Flow Index	Volume-weighted RSI measuring buying/selling pressure
STOCH	Stochastic Oscillator	Compares closing price to price range over N periods
ROC	Rate of Change	Percentage price change over N periods
ADX	Average Directional Index	Measures trend strength regardless of direction
CCI	Commodity Channel Index	Measures price deviation from statistical mean
MACDH	MACD Histogram	Difference between MACD line and signal line

Note

Each feature has configurable parameters (periods, smoothing, etc.) accessible through the feature settings panel. The default parameters are reasonable starting points.

ML Parameters

Parameter	Description	Guidance
Neighbors (K)	Number of nearest neighbors to consider	Lower K (3-5) = more responsive but noisier. Higher K (8-15) = smoother but slower
Prediction Days	How far ahead the model predicts	Shorter (1-3) for day trading. Longer (5-10) for swing trading
Max Bars Back	Historical bars used for pattern matching	More bars = more patterns but slower computation. 2000-5000 is typical

Kernel Configuration

The kernel determines how the model measures similarity between current and historical patterns. Different kernels emphasize different aspects of the distance calculation.

Kernel	Description
RQ (Rational Quadratic)	Flexible kernel that adapts to multiple length scales. Good general-purpose choice.
RQ + Gaussian	Combines RQ with Gaussian kernel for both local and global pattern matching.
None	Raw distance-based matching without kernel transformation.

Exit Strategies

Exit strategies determine when the model suggests closing a position. Multiple exit methods can be active simultaneously; the first triggered exit closes the position.

Strategy	Description
Time-Based	Exit after N bars regardless of profit/loss
Opposite Signal	Exit when the model generates a contrary signal
Stop Loss	Exit if price drops below a fixed percentage threshold
ATR-Based	Exit using Average True Range for dynamic stop levels
Trailing Stop	Exit using a trailing stop that follows price upward
Take Profit	Exit when a target profit percentage is reached

Entry Filters

Entry filters add additional conditions that must be met before a signal is acted upon. They help reduce false signals by requiring confirmation from other indicators.

●Volume Filter: Requires minimum volume threshold
●RSI Filter: Requires RSI within a specified range
●Volatility Filter: Requires volatility within acceptable bounds
●Momentum Filter: Requires positive momentum confirmation
●ADX Filter: Requires minimum trend strength
●Trend / EMA Filter: Requires price alignment with moving averages
●SMA Filter: Requires price above/below a Simple Moving Average
●Regime Filter: Requires specific market regime conditions

Advanced Settings

Core machine learning parameters including distance methods, weighted voting, ensemble settings, adaptive K, LOF outlier detection, and probability calibration.

Exits

Configure exit strategy combinations, stop loss types, trailing parameters, take profit levels, and time-based exits.

Filters

Set entry filter thresholds for volume, momentum, volatility, trend, and regime conditions.

Models

Select and configure distance metrics (Euclidean, Manhattan, Chebyshev, Cosine, Minkowski, Lorentzian, Mahalanobis).

Backtest Results

After running an analysis, the results panel shows detailed backtest statistics including total trades, win rate, profit factor, maximum drawdown, and equity curve. Use these results to evaluate configuration effectiveness before relying on live signals.

Metric	Description
Total Trades	Number of trades generated in the backtest period
Win Rate	Percentage of trades that were profitable
Profit Factor	Gross profit divided by gross loss (above 1.0 = profitable)
Max Drawdown	Largest peak-to-trough decline in the backtest
Sharpe Ratio	Risk-adjusted return metric
Avg Win / Avg Loss	Average gain on winning trades vs average loss on losing trades

Best Practices

●Start with default settings before customizing
●Change one parameter at a time to understand its impact
●Ensure backtest has at least 20+ trades for statistical significance
●Watch for overfitting: extremely high win rates on few trades may not generalize
●Test configurations across different market conditions and time periods
●Save successful configurations for reuse

Warning

Backtest results reflect historical performance only. Overfitted configurations that look excellent in backtests may perform poorly on new data. Always validate across multiple time periods and market conditions.

Common Configuration Templates

Template	Settings	Best For
Conservative Swing	K=8, Pred Days=5, RSI+MACD+ADX features, RSI+Trend filters, Stop 5%	Lower-frequency, higher-confidence swing trades
Aggressive Day	K=3, Pred Days=1, RSI+WT+STOCH features, Volume filter only, Stop 2%	High-frequency signals for active day traders
Momentum Follow	K=5, Pred Days=3, ROC+ADX+MACDH features, ADX+Momentum filters, Trailing 3%	Trending stocks with strong directional moves
Mean Reversion	K=10, Pred Days=5, RSI+MFI+CCI features, RSI+Volatility filters, Take Profit 3%	Overbought/oversold bounce setups

Tip

These templates are starting points. Every stock has different volatility and behavior characteristics. Run the backtest on your specific symbol to see which template works best, then fine-tune from there.

Combining with Other Tools

●Configure ALEEP here, then visualize signals on AI Charts for an integrated visual experience
●Use Market Regimes to inform your filter settings — enable Regime Filter during choppy markets
●Cross-reference ALEEP Config backtest results against the same stock's Composite Score for fundamental validation
●Apply signals from ALEEP Config to stocks identified through the Screener or Pattern Signals for multi-layer analysis

This platform provides data and analysis tools for educational and informational purposes only. Nothing on this platform constitutes financial advice, investment recommendations, or solicitation to buy or sell any securities. Always conduct your own research and consult with a qualified financial advisor before making investment decisions.