Statistical Arbitrage Pairs

Entry Logic

• Long the underperforming stock and short the outperforming stock of a correlated pair when the spread diverges by 2 standard deviations from the mean.
• Confirmation requires the spread to start mean-reverting.
• Timeframe is 15-minute chart.
• Location context is a divergence from the historical correlation.
• Market condition should be relatively stable, without major news affecting either stock.

Exit Logic

• Profit target is when the spread reverts to the mean.
• No scaling out.
• Trailing stop is not used.
• Exit on signal failure if the spread widens to 3 standard deviations.
• Exit on opposite signal is not applicable.
• Exit on time expiration after 5 trading days.
• Exit on momentum loss if the spread stalls for more than one day.

Stop Loss Structure

• Hard stop is a 3 standard deviation move against the position.
• Soft stop is a fundamental change in one of the companies.
• Maximum dollar loss is $150 per trade.
• Maximum percent loss is 1.5% of account equity.
• Structural stop is not applicable.

Risk Management Framework

• Risk per trade is 0.75% of account equity.
• Maximum daily loss limit is 2.25% of account equity.
• Maximum weekly loss limit is 4.5% of account equity.
• Maximum drawdown is 18%.
• Risk-reward ratio requirement is a minimum of 1:1.

Position Sizing Model

• Sizing is dollar-neutral, with equal dollar amounts in the long and short positions.
• Volatility adjustment is used to balance the beta of the two stocks.
• Conviction sizing is not used.
• No scaling in.
• No scaling out.

Trade Filtering

• Avoid pairs with low historical correlation (below 0.8).
• Only trade pairs within the same sector.
• Avoid holding pairs through earnings announcements.
• Time of day restrictions are not a primary concern.
• Avoid pairs where one stock is subject to a takeover bid.

Context Framework

• Trend direction of the overall market is not a primary concern.
• VWAP and moving average relationships are not used.
• Range location is based on the standard deviation of the spread.
• Higher timeframe alignment is not applicable.

Trade Management Rules

• Do not adjust the stop loss.
• No scaling out.
• Do not add to the position.
• The trade is managed as a single unit until the profit target or stop loss is hit.

Time Rules

• Optimal trading window is during regular market hours.
• Avoid the first and last 15 minutes of the day.
• Session notes are not applicable.

Setup Classification

• A+ setup: 2.5 standard deviation divergence with a high historical correlation.
• A setup: 2.0 standard deviation divergence.
• B setup: 1.5 standard deviation divergence.
• C setup: Less than 1.5 standard deviation divergence.

Market Selection Criteria

• Instruments are pairs of highly correlated stocks.
• Minimum daily volume of 2 million shares for each stock.
• Volatility of the spread should be stable.

Statistical Edge Metrics

• Expected win rate is 70%.
• Average win is 1R.
• Average loss is 1R.
• Profit factor is 2.3.
• Expectancy per trade is +0.4R.

Failure Conditions

• Strategy fails if the correlation between the pair breaks down.
• A fundamental change in one of the companies can cause a permanent divergence.
• Avoid during times of high market stress.

Psychological Rules

• Must be patient and wait for the statistical signal.
• Avoid closing the trade early before the spread reverts to the mean.
• Trust the statistical model and do not interfere with the trade.

Advanced Components

• Market regime detection is not used.
• Volatility filter monitors the VIX for signs of market stress.
• Correlation filter is the core of the strategy.
• Multi-timeframe alignment is not applicable.

Location

• Strongest in stable, range-bound markets.
• Weakest during market panics or when a sector is undergoing a major shift.
• The strategy's success depends on the stability of the pair's relationship.