Direct Answer
The linear regression intercept is the value at which the least-squares line crosses the vertical axis at the start of the fitting window. It is half of the description of the fitted line, with the slope being the other half. Its value depends on where x is counted from, so it is meaningful only together with the slope and the x convention.
Linear Regression Intercept: Formula, Meaning, Signals, Examples and How to Use It
How Is Linear Regression Intercept Calculated?
The linear regression intercept is the value at which the least-squares line crosses the vertical axis at the start of the fitting window. It is half of the description of the fitted line, with the slope being the other half.
Formula: intercept = (sum(y) - slope * sum(x)) / N, which equals mean(y) - slope * mean(x)
Inputs:
- Close prices
- Regression length N
- Fitted slope
Worked example. With a total of 60, an x total of 15, a slope of 0.9 and 5 points, the intercept is 9.30. The inputs are illustrative numbers, not market data.
Reading it. Its value depends on where x is counted from, so it is meaningful only together with the slope and the x convention. It is mainly an input for the regression value and for forecasts.
Limits. It extrapolates back to x = 0, which lies outside the data, so small slope errors move it a lot. Used alone it carries little information about the market.
What is Linear Regression Intercept?
The linear regression intercept is the value at which the least-squares line crosses the vertical axis at the start of the fitting window. It is half of the description of the fitted line, with the slope being the other half.
How to use Linear Regression Intercept
Its value depends on where x is counted from, so it is meaningful only together with the slope and the x convention. It is mainly an input for the regression value and for forecasts. A reading alone is never a buy or sell signal; it needs context from price structure, volume, volatility and the broader market regime.
Strengths
- Together with the slope it fully specifies the fitted line, so the intercept completes a two-parameter description of the window.
- Because it equals the mean of y minus slope times the mean of x, it can be checked from two averages and the slope.
- Reproducible when the formula and data convention are fixed.
Weaknesses and limitations
- Its value depends on how x is numbered (starting at 0 or at 1), so the same fit gives different intercepts under different conventions.
- Because it is anchored at the first bar of the window, it moves each time the window advances even when the slope does not.
- Correlated indicators can create false confidence by repeating the same underlying information.
Editorial and risk note
This page is educational content, not individualized investment advice. It should not imply guaranteed prediction, accuracy or outperformance.
Frequently Asked Questions
Is Linear Regression Intercept a buy or sell signal?
No. Its value depends on where x is counted from, so it is meaningful only together with the slope and the x convention. It extrapolates back to x = 0, which lies outside the data, so small slope errors move it a lot. A trade built on it still needs a trigger, an invalidation rule, a position size and realistic execution assumptions.
What is the best setting for Linear Regression Intercept?
Linear Regression Intercept takes these inputs: close prices; regression length N; fitted slope. There is no universal best setting for the tunable ones, so test them across instruments, regimes and out-of-sample periods and favor stable parameter regions over one historical winner.
Can Linear Regression Intercept be used by itself?
It extrapolates back to x = 0, which lies outside the data, so small slope errors move it a lot. Used alone it carries little information about the market. For that reason it is read next to measures of direction, participation, valuation, liquidity and risk, not alone.
Does Linear Regression Intercept work on every timeframe?
The calculation (intercept = (sum(y) - slope * sum(x)) / N, which equals mean(y) - slope * mean(x)) can be run on any bar size, but behavior is not portable: a reading on a five-minute chart describes a very different market window from the same reading on a daily chart. Validate the exact timeframe before relying on it.
Why does Linear Regression Intercept give false signals?
Its value depends on how x is numbered (starting at 0 or at 1), so the same fit gives different intercepts under different conventions. Because it is anchored at the first bar of the window, it moves each time the window advances even when the slope does not. Lag, noise and regime changes add to this, because market participants react to new information after the reading is calculated.
How should Linear Regression Intercept be backtested?
Rebuild Linear Regression Intercept from close prices using point-in-time data, write the rules before testing, include delisted securities where relevant, model realistic fills and costs, reserve a validation sample and compare with a simpler baseline.