Definitions of mathematical and statistical terms used throughout Liquic's UI and data pipeline
Belief-Weighted Aggregated Probabilities. BWAPs are probability distributions created by aggregating market beliefs from prediction markets (like Kalshi and Polymarket), weighted by trading volume and open interest. These distributions represent the market's collective view of future outcomes, enabling AI agents to simulate, evaluate, and act under uncertainty.
Home page description, Beliefs page, and throughout the data pipeline where prediction market data is aggregated into probability distributions.
A mathematical function that describes the likelihood of different outcomes. In Liquic, probability distributions represent market beliefs about future prices or events, showing not just a single prediction but the full range of possible outcomes with their associated probabilities.
Beliefs page (yearly low distributions), SIPmath exports, and simulation models throughout the application.
A quantile is a value below which a given percentage of observations fall. For example, the 90th percentile (p90) is the value below which 90% of outcomes occur. Percentiles are quantiles expressed as percentages (p50 = 50th percentile = median).
Paths page (fan bands showing p10, p50, p90, etc.), Beliefs page (querying prices at specific probability levels), and simulation results.
A function that gives the probability that a random variable is less than or equal to a given value. The CDF maps any value to its cumulative probability, showing how probability accumulates from the minimum to that point.
Metalog workbench, Beliefs page (converting between prices and probabilities), and SIPmath distribution representations.
A function that describes the relative likelihood of a random variable taking on a given value. The PDF shows the "density" of probability at each point, where higher values indicate more likely outcomes. The area under the PDF curve equals 1 (100% probability).
Metalog workbench visualizations and distribution fitting diagnostics.
A flexible probability distribution that can represent a wide variety of shapes using a small number of parameters (coefficients). Metalog distributions are particularly useful for fitting to market data because they can capture skewness, fat tails, and other non-normal characteristics common in financial markets.
Beliefs page (fitted distributions from prediction markets), Paths page (simulation models), Metalog workbench, and SIPmath exports.
The multiplier applied to your position size, expressed as "x" (e.g., 10x leverage). With 10x leverage, a 10% price move results in a 100% gain or loss. Higher leverage increases both potential returns and liquidation risk.
Cliff page (leverage selector), Paths page (simulation inputs), and liquidation probability calculations.
The percentage price move required to reach your liquidation price. For a LONG position with 10x leverage, liquidation occurs if price drops by 10% (distance = 1/leverage). This represents how close your position is to being automatically closed.
Cliff page (showing liquidation probability based on distance), Paths page (simulating paths that may touch liquidation price).
The probability that a leveraged position will be liquidated within a given time horizon (e.g., 24 hours). In Liquic, liquidation is modeled as a "touch" event—if price touches the liquidation price at any point, the position is closed. This probability is calculated using historical price movement patterns.
Cliff page (24-hour liquidation probability), Paths page (simulated paths showing liquidation events), and survival analysis.
A single sampled trajectory of future price movements, generated by randomly drawing from probability distributions fitted to historical data. Each path represents one possible future scenario, showing how price might evolve over time.
Paths page (individual path visualizations), simulation results showing multiple possible futures, and fan band calculations.
Visual representations of uncertainty showing the range of possible outcomes at different probability levels. Fan bands typically display percentiles (e.g., p10–p90) as shaded regions, with the median (p50) as a central line. Wider bands indicate greater uncertainty.
Paths page (fan chart showing p10, p50, p90 bands over time), simulation visualization, and risk analysis displays.
OHLC (Open, High, Low, Close) data represents price movements over a time period. A "candle" is a visual representation of OHLC data. The return is the percentage change from open to close. Liquic uses historical OHLC data to model price movements, including intraday wicks (high/low) that can trigger liquidations.
Cliff page (down/up move series from candles), Paths page (simulation inputs), and historical data analysis throughout the application.