Struct PriceFeatures 

pub struct PriceFeatures {
    pub log_returns: Vec<f64>,
    pub realized_volatility: f64,
    pub momentum: f64,
    pub rsi: f64,
    pub macd_signal: f64,
    pub bollinger_position: f64,
}

Price-derived features computed from a close-price series.

Example

use fin_primitives::ml_features::PriceFeatures;

let closes = vec![100.0, 101.0, 102.0, 101.5, 103.0];
let pf = PriceFeatures::compute(&closes, 3).unwrap();
assert!(pf.log_returns.len() == 4);

Fields

log_returns: Vec<f64>

Log returns: ln(close_t / close_{t-1}).

realized_volatility: f64

Realized volatility: rolling std-dev of log returns over the last window periods.

momentum: f64

n-period momentum: close[-1] / close[-n-1] - 1.

rsi: f64

RSI over the last window periods (0–100).

macd_signal: f64

MACD signal line: EMA(12) − EMA(26) of closes.

bollinger_position: f64

Bollinger Band position: (close - lower) / (upper - lower), in [0, 1].

Implementations

impl PriceFeatures

pub fn compute(closes: &[f64], window: usize) -> Result<Self, FinError>

Compute price features from a close-price series.

Requires at least window + 1 observations for all features.

Errors

• FinError::InsufficientData if fewer than window + 2 closes.
• FinError::InvalidPeriod if window == 0.

Trait Implementations

impl Clone for PriceFeatures

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for PriceFeatures

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.