Volatility Clustering Algorithm

Entry Logic

• A GARCH model is used to forecast volatility.
• A long entry is triggered when volatility is forecasted to be low, and a breakout signal occurs.
• A short entry is triggered under the same low-volatility forecast, but with a breakdown signal.
• Confirmation is a significant increase in volume on the breakout or breakdown.
• The timeframe is daily.
• The location context is a period of low volatility.
• The market condition is a consolidating market that is expected to transition to a trending market.

Exit Logic

• The exit is triggered when volatility is forecasted to be high, and a reversal signal occurs.

Stop Loss Structure

• The stop loss is placed at a level that invalidates the breakout or breakdown signal.

Risk Management Framework

• Risk management rules are applied to the trades generated by the volatility clustering algorithm.

Position Sizing Model

• Position sizing is adjusted based on the volatility forecast.

Trade Filtering

• Trades are filtered based on the volatility forecast.

Context Framework

• The volatility forecast provides the context for the market.

Trade Management Rules

• The trade is managed based on the evolution of volatility.

Time Rules

• The strategy can be applied at any time.

Setup Classification

• The strength of the setup is determined by the strength of the volatility forecast and the quality of the breakout or breakdown signal.

Market Selection Criteria

• The strategy is best suited for markets that exhibit volatility clustering.

Statistical Edge Metrics

• The edge is determined by backtesting the strategy.

Failure Conditions

• The strategy can fail if the volatility forecast is inaccurate.

Psychological Rules

• The main challenge is to be patient and wait for the right volatility conditions to enter a trade.

Advanced Components

• A variety of GARCH models can be used, such as EGARCH and GJR-GARCH.

Location

• The strategy is most effective in markets that exhibit clear periods of low and high volatility.