LinearRegressionIntercept — Linear Regression Intercept

The fitted value at the window’s first bar: where the regression line starts. Paired with the slope it reconstructs the whole line, and the gap between intercept and current price is how far the window has travelled along its own trend.

Signature

LinearRegressionIntercept(period, on=<source>)

Argument	Type	Required	Notes
period	int Param or literal	Yes	Regression window. Default 14, bounds 2–400.
on=	symbol, OHLCV field, or indicator	No	Source. Defaults to the trade symbol’s close.

Formula

$$ \text{intercept} = \frac{\sum C - \text{slope}\sum i}{n} $$

with i = 0 at the oldest bar in the window.

Warmup

period bars; tracks the close until then.

Reading the value

lri.value
self.trade.close - lri.value      # total move along the fitted line

Example

class TrendTravel(Strategy):
    """Enter when the window has travelled far from where its trend began."""
    trade = Symbol()
    sizing = FixedCapital()

    def execute(self):
        lri = LinearRegressionIntercept(20, on=self.trade)
        if self.trade.close > lri.value * 1.03 and self.trade.is_flat:
            self.buy(self.trade, reason="Trend has covered ground")
        if self.trade.close < lri.value and self.trade.is_long:
            self.sell(self.trade, self.trade.position, reason="Back to the start of the fit")

Notes

• TA-Lib LINEARREG_INTERCEPT.
• Proven by an assert-via-trade run on SPY daily (matched a strategy-side OLS intercept to 1e-3 on every bar; zero alarm trades).
• The intercept is anchored to the oldest bar in the window, so it moves as the window rolls — it is not a fixed anchor.