Crate motosan_ta_math 

motosan-ta-math

Pure mathematics and analytics layer for the motosan-ta workspace.

Key modules

• indicators — technical indicators (moving averages, momentum, volatility, volume)
• patterns — chart pattern detection (VCP, Cup & Handle, Double Bottom, harmonic)
• risk — stop-loss rules, position sizing, Kelly criterion, VaR, drawdown
• portfolio — Markowitz mean-variance optimization (MinVariance, MaxSharpe, RiskParity)
• position — fill-based position tracker with VWAP cost basis and realized PnL
• correlation — Pearson correlation and rolling correlation matrix
• factor — OLS multi-factor regression and variance decomposition
• microstructure — Amihud illiquidity, Kyle’s lambda, Roll bid-ask spread
• pairs — pairs trading: OLS hedge ratio, ADF test, cointegration, spread z-score
• pnl_attribution — trade PnL decomposition into direction, timing, slippage, and fees
• models — shared domain types (Quote, TrendStage, EntrySignal, etc.)

Example

use motosan_ta_math::indicators::{ema, sma};

let prices = vec![10.0, 11.0, 12.0, 11.5, 13.0];
let ema_values = ema(&prices, 3);
let sma_values = sma(&prices, 3);

Modules

correlation

Pearson correlation and rolling correlation matrix.

error

Re-exports TaError from motosan-ta-primitives as the single error type for this crate.

factor

Multi-factor OLS regression and variance decomposition.

indicators

microstructure

Market microstructure metrics for liquidity and price-impact estimation.

models

Shared domain models used across ta-math and ta-batch.

pairs

Pairs trading utilities: hedge ratio, cointegration testing, and spread z-score tracking.

patterns

pnl_attribution

Trade PnL decomposition into direction, timing, slippage, and fee components.

portfolio

Markowitz-style portfolio optimization with four objectives.

position

Fill-based position tracker with VWAP cost basis and realized/unrealized PnL.

risk

Functions

ping