Part of the Maths & Statistics suite · 235 calculators

Quadratic Regression Calculator

Fit a parabola y = ax² + bx + c to your data — for relationships that rise then fall, or curve away from a straight line.

Quadratic regression is still least squares; the model just includes an x² term.

Results update as you type
Results
a (x² coefficient)
-0.880952
b (x coefficient)
c (constant)
Equation
Vertex — x
Vertex — y
Shape
Prediction
Reviewed September 2026. Pure mathematics: the result does not depend on where you are. Terminology follows the national curriculum (maths, brackets, decimal point).
No account required · Google Analytics off unless allowedCalculator arithmetic runs in your browserResults update as you type
All calculations run 100% in your browser. The calculator code does not submit your figures to GlobalCalc to obtain a result.
About quadratic regression

How the quadratic regression calculator works

Quadratic regression is still least squares; the model just includes an x² term. The three normal equations are solved directly by Gaussian elimination on a 3×3 system.

The vertex, at x = −b/2a, is usually the point of interest: the dose that maximises response, the speed that minimises fuel use, the temperature at peak yield.

Formula: y = ax² + bx + c; vertex at x = −b / 2a

Worked examples

Inputsa (x² coefficient)Note
A clean parabola-0.880952peaks near x = 4
An exact fit2y = 2x² + 0x + 1
A straight line0a ≈ 0 — a quadratic is unnecessary

Frequently asked questions

When should I fit a quadratic?

When the relationship curves — rises then falls, or accelerates. A residual plot from a straight-line fit showing a clear arc is the usual sign.

What is the vertex for?

It locates the turning point: the optimum dose, speed or temperature. It is often the whole reason for fitting a curve.

Does a quadratic always fit better?

Its R² can never be lower than a straight line's, because the line is a special case. That is why a higher R² alone does not justify the extra term.

Can I extrapolate from it?

Very cautiously. A parabola shoots off to infinity in both directions, so predictions outside your data range go wrong fast.

What if a is nearly zero?

The curvature is negligible and a straight line is the better model — simpler, and it will predict better out of sample.

Where these figures come from

Last checked: September 2026. Formulas are fixed by mathematics and do not change with tax years or regulations.