Inspect the models, not just the marketing.
Every public built-in LargeQuant model exposes its family, version, runtime, input contract, output contract and default parameters. Finance-specific private/internal model brands are intentionally excluded from this public registry.
Signal Strategy Backtester
lq.backtest.signal-strategyLagged-signal strategy backtest with look-ahead guard, transaction costs, turnover, drawdown and risk-adjusted performance metrics.
Streaming Time-Series Quality Profile
lq.data.time-series-profileProfiles ordered numerical streams for timestamp monotonicity, duplicates, cadence gaps and robust MAD outliers before downstream quantitative research.
Evidence-Backed Adaptive Model Router
lq.decision.adaptive-routerSelects a champion or adaptive ensemble from candidate performance histories using recency weighting, optional regime conditioning, minimum-evidence guards and switching margins.
Weighted Sum
lq.math.weighted-sumDeterministic weighted aggregation with explicit weights and reproducible arithmetic.
VAR(1) Multivariate Econometric Model
lq.econometrics.var1Multivariate VAR(1) estimation by OLS with residual covariance, per-series diagnostics and recursive multi-step forecasts.
Adaptive Exponential-Weights Ensemble
lq.ensemble.exponential-weightsRecency-aware exponential weighting over candidate loss histories with deterministic champion selection and effective-model-count diagnostics.
Prequential Forecast Evaluation
lq.evaluation.prequential-forecastOnline one-step-ahead evaluation of naive, AR(1) and linear-trend forecasters using only information available before each realized observation.
AR(1) Residual-Bootstrap Forecast
lq.forecasting.ar1-residual-bootstrapSeeded empirical residual bootstrap around an AR(1) process, producing distributional forecasts, quantile intervals and reproducible sample paths.
Linear Trend Forecast
lq.forecasting.linear-trendOLS time-trend forecast with residual uncertainty intervals. Quantitative output only; no language generation.
Quantitative Memory Performance Profile
lq.memory.performance-profileSummarizes signed quantitative-memory observations by candidate, recency and current regime to produce reproducible performance profiles.
Page-Hinkley Sequential Drift Detector
lq.monitoring.page-hinkleySequential Page-Hinkley drift detector for realized errors, residuals or other numerical monitoring streams with explicit alarm indices and drift statistics.
Two-Sided CUSUM Drift Detector
lq.monitoring.two-sided-cusumSequential two-sided cumulative-sum detector for upward or downward shifts in realized error, residual, risk or other quantitative streams.
Capacity allocation
lq.optimization.capacity-allocationExact bounded integer capacity allocation, with a first-fit baseline and scoped assessment of a supplied allocation.
Constrained Mean-Variance Optimizer
lq.optimization.mean-varianceLong-only capped portfolio optimisation using projected-gradient optimisation on expected returns and covariance.
Structural Break Detector
lq.regime.structural-breakExhaustive two-regime least-squares mean-shift scan with fit-improvement and standardized-shift diagnostics.
Two-State Gaussian HMM Regime Model
lq.regime.gaussian-hmm2Deterministic two-state Gaussian hidden Markov model estimated with scaled Baum-Welch EM, emitting state probabilities, transition persistence, regime separation and current latent-state classification.
Autonomous Challenger Research Cycle
lq.research.challenger-cyclePrequential champion/challenger evaluation with sequential error-drift monitoring and explicit evidence-based recommendation to hold, challenge or initiate fresh research.
Rolling Probabilistic Forecast Calibration
lq.research.rolling-forecast-calibrationOut-of-sample rolling AR(1) residual-bootstrap calibration with empirical interval coverage, width, interval score and explicit look-ahead guard.
Walk-Forward AR(1) Research Study
lq.research.walk-forward-ar1Out-of-sample one-step AR(1) evaluation against a naive last-value baseline with explicit look-ahead guard and LQBench skill metrics.
Walk-Forward Forecast Model Competition
lq.research.forecast-model-competitionOut-of-sample competition among naive, AR(1) and linear-trend forecasters with deterministic winner selection or inverse-RMSE ensemble weighting.
Dynamic EWMA Covariance & Correlation
lq.risk.dynamic-ewma-covarianceMultivariate exponentially weighted covariance engine with current covariance, correlation, volatility and a bounded covariance-path tail for dynamic risk monitoring.
Historical VaR & Expected Shortfall
lq.risk.historical-varEmpirical historical Value at Risk, CVaR/Expected Shortfall, volatility and drawdown from a return series.
Probabilistic Portfolio Risk Surface
lq.risk.probabilistic-surfaceSeeded multivariate Monte Carlo risk surface across multiple horizons and confidence levels, emitting Value at Risk, Expected Shortfall and P&L distribution summaries.
Portfolio Stress Scenario Engine
lq.scenario.portfolio-stressDeterministic delta-gamma portfolio stress revaluation across named shock scenarios with worst/best and tail-loss summaries.
Correlated Multi-Asset GBM Simulator
lq.simulation.correlated-gbmSeeded multi-asset geometric Brownian motion using a validated covariance matrix and Cholesky-correlated innovations.
Geometric Brownian Motion Simulator
lq.simulation.geometric-brownian-motionSeeded stochastic path simulation with reproducible terminal distributions and sample paths.
Local-Level State-Space Model
lq.state-space.local-levelLocal-level Kalman filter with transparent process/observation variance assumptions, innovations and probabilistic state forecasts.
Descriptive Statistics
lq.statistics.descriptiveAuditable descriptive statistics for a numeric series, including dispersion and interpolated percentiles.
Linear Regression
lq.statistics.linear-regressionOrdinary least squares regression for one explanatory series with explicit fit diagnostics.
Pearson Correlation
lq.statistics.pearson-correlationDeterministic Pearson product-moment correlation for two aligned numeric series.
AR(1) Time-Series Forecast
lq.timeseries.ar1Autoregressive AR(1) estimation and recursive forecasting with residual uncertainty propagation and stationarity diagnostics.
EWMA Volatility Model
lq.volatility.ewmaExponentially weighted volatility estimator with explicit decay, annualization and volatility-path diagnostics.
Variance-Targeted GARCH(1,1)
lq.volatility.garch11GARCH(1,1) conditional volatility model using variance targeting and deterministic Gaussian quasi-likelihood search with persistence diagnostics.