LinearRegressionCurve — Linear Regression Curve

The fitted regression value plotted at each bar. Numerically identical to LSMA; it ships under both names because charting platforms use both, and a strategy reading “the regression curve” should find it by that name.

Signature

LinearRegressionCurve(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

$$ \text{value}_t = \text{intercept}_t + \text{slope}_t,(n-1) $$

recomputed on each bar’s window.

Warmup

period bars; the raw close until the window fills.

Reading the value

lrc.value
self.trade.close > lrc.value

Example

class RegressionBaseline(Strategy):
    """Trade around the regression curve as a fair-value line."""
    trade = Symbol()
    sizing = FixedCapital()

    def execute(self):
        lrc = LinearRegressionCurve(20, on=self.trade)
        if self.trade.close < lrc.value * 0.97 and self.trade.is_flat:
            self.buy(self.trade, reason="Below the fit")
        if self.trade.close > lrc.value and self.trade.is_long:
            self.sell(self.trade, self.trade.position, reason="Back at the fit")

Notes

• Same computation as LSMA (different default period). Pick one name and stay with it inside a strategy.
• Proven by an assert-via-trade run on SPY daily (matched a strategy-side OLS reference to 1e-3 on every bar; zero alarm trades).
• For the slope or fit quality, use LinearRegressionSlope or LinearRegression.