Quant Reference

Measures the magnitude of recent price changes to evaluate overbought or oversold conditions on a scale of 0–100. Readings above 70 suggest overbought, below 30 suggest oversold. Uses Wilder's smoothing method with a 14-period lookback.

Shows the relationship between two moving averages of a security's price. The MACD line is the 12-period EMA minus the 26-period EMA. The signal line is a 9-period EMA of the MACD. Crossovers indicate trend changes.

A volatility band plotted 2 standard deviations above and below a 20-period SMA. When price touches the upper band, the stock is overextended. Touching the lower band suggests a potential bounce. Band squeeze (narrowing) precedes breakout.

The arithmetic mean of closing prices over a specified period. QuantBrainAI computes SMA(50) and SMA(200). A Golden Cross (SMA50 crossing above SMA200) signals bullish momentum. A Death Cross (SMA50 crossing below) signals bearish reversal.

The ratio of cumulative (price × volume) to cumulative volume over a 20-period window. Institutional traders use VWAP to assess whether they bought below or above the average price. Price above VWAP is bullish; below is bearish.

Measures market volatility by decomposing the entire range of an asset price for a given period. Calculated as the SMA of the True Range (max of: current high − low, high − previous close, low − previous close) over 14 periods.

A cumulative volume indicator that adds volume on up days and subtracts on down days. OBV divergence — price moving up while OBV moves down — signals weakening bullish momentum and potential reversal.

Compares current trading volume to the 20-day average. Volume spikes above 1.5× indicate unusual activity — often preceding breakouts or reversals. Volume drops below 0.5× suggest low conviction or consolidation.

High — RSI extreme overbought/oversold, Bollinger lower band touch, volume spike >2×, Death cross, OBV divergence. These signal regime changes, not noise.

Medium — RSI overbought/oversold, MACD crossovers, Bollinger upper touch, Golden cross, volume drop, ATR spike. Standard technical alerts.

Low — MA support confirmation, VWAP rejection. Contextual signals used for confirmation.

When 3+ high-severity signals converge on the same direction, the system increases conviction weighting. QuantBrainAI's engine applies non-linear amplification — convergence amplifies the composite score more than the raw signal sum would suggest. This prevents linear additive bias.

The difference between a player's baseline score (standalone capabilities) and their equilibrium score (best response in the competitive matrix). A positive divergence means the player benefits disproportionately from market structure — their position is stronger than raw fundamentals suggest. Negative divergence indicates structural disadvantage.

15 independent strategy agents score each ticker across dimensions: value (P/E, EV/EBITDA, discounted cash flow), momentum (price trend, volume acceleration), quality (ROE, profit margins, debt ratios), volatility (beta, ATR percentile), trend (SMA alignment, MACD regime), and sentiment (social media buzz, option flow). Each agent votes independently — the ensemble reduces individual model bias.

Three-layer quantification of data disadvantage against well-resourced institutional competitors. Scores 13 external-data dimensions on freshness, exclusivity, and price-reflection. Computes a weighted composite disadvantage (currently ~29%), applies non-linear cross-factor amplification when 3+ positive or negative signals converge, and widens tail-risk scenario probabilities proportionally.

Game Theory Equilibrium (35%) — Nash divergence, ecosystem force, top influencer power
Technical Indicators (30%) — RSI, MACD, Bollinger, OBV, volume, MA cross
Signal Detection (20%) — severity-filtered signal impact with convergence amplification
Fundamental Data (15%) — P/E ratio, market cap, 52-week range, analyst consensus

Buy (≥80) — Multiple pipelines confirm bullish thesis. Game theory shows favorable equilibrium. Technicals and fundamentals align.

Hold (50–79) — Mixed signals. Some pipelines bullish, others cautious. No strong conviction either direction.

Sell (<50) — Multiple pipelines bearish. Structural disadvantage, deteriorating technicals, negative signals.

When pipeline data is missing (stale game-state, expired external context), the remaining components still contribute. Missing pipelines are treated as neutral (score contribution = 0), not as negative. The composite trends neutral rather than producing false conviction from partial data.

A zero-shot time-series foundation model pretrained on 100 billion+ time points across diverse domains. QuantBrainAI uses it for price forecasting without fine-tuning — it generates point forecasts, 80% confidence interval bands, trend direction, and anomaly flags. Runs entirely CPU-only at ~1.5 GB RAM. Daily cached per ticker.

15 independent agents score each ticker separately. The ensemble averages individual predictions using weighted voting (weights learned from historical accuracy per agent per market regime). This reduces overfitting to any single model and adapts to changing market conditions.

The excess return of an investment relative to the return of a benchmark index. Positive alpha indicates a portfolio has outperformed on a risk-adjusted basis. The "holy grail" of quantitative finance — every strategy is ultimately measured by its alpha generation.

Measures the volatility of a stock relative to the overall market. A beta of 1.5 means the stock is 50% more volatile than the market — it tends to move 1.5% for every 1% market move. Semiconductor stocks typically have high beta (>1.2).

The average return earned in excess of the risk-free rate per unit of volatility. A Sharpe ratio above 1.0 is considered good, above 2.0 excellent. Quant strategies aim for high Sharpe ratios — not just high returns, but consistent returns with low drawdown.

Like the Sharpe ratio but penalizes only downside volatility. Useful for evaluating strategies where upside volatility is acceptable (or desirable). A Sortino ratio of 3.0+ is excellent for long-term strategies.

The maximum observed loss from a peak to a trough of a portfolio, before a new peak is achieved. The key metric for risk management — a strategy returning 30% is meaningless if it endured a 60% drawdown to get there. Quant strategies target max drawdown below 15–20%.

Simulating a trading strategy on historical data to evaluate its performance. QuantBrainAI backtests all signal combinations across multiple market regimes (bull, bear, high-volatility, low-volatility) before deploying to production. Key pitfalls: overfitting, survivorship bias, look-ahead bias.

Harry Markowitz's Modern Portfolio Theory — selecting the mix of assets that maximizes expected return for a given level of risk. The efficient frontier represents portfolios offering the highest expected return for each risk level. QuantBrainAI applies this across semiconductor, cloud, and infrastructure holdings.