Least Squares Approximation

S_r = \sum_{i=1}^n e_i^2= \sum_{i=1}^{n} (y_{i,\text{measured}} - y_{i,\text{model}})^2

Linear Regression

The least squares approach that involves determining the best approximating line.

Let the best least squares line to a collection of data points ${(x_i,y_i)_{i=1}^n}$ be:

y = a_1 x + a_0

$a_1$ and $a_0$ can be found by solving:

a_1 \sum_{i=1}^n {x_i^2} + a_0\sum_{i=1}^n {x_i} = \sum_{i=1}^n {x_i y_i}

a_1 \sum_{i=1}^n {x_i} + a_0\sum_{i=1}^n {1} = \sum_{i=1}^n {y_i}

The least squares approach that involves determining the best approximating polynomial.

Let the best least squares polynomial to a collection of data points ${(x_i,y_i)_{i=1}^m}$ be:

P_n(x) = \sum_{i=1}^{n} a_i x^i

Here $n \lt m-1$ .

The constants $a_i$ can be found by subtituting the known data points in the polynomial.

\sum_{k=0}^n \left\{a_k \sum_{i=1}^m x_i^{j+k} \right\} = \sum_{i=1}^m y_i x_i^{j} \;\;\text{where}\; j = 0,1,\dots,n

Wolfram Mathematica notebook

The summary notebook defines every function used in this module, with documentation and no examples, as a starting point for your own problems.

Written by

September 29, 2026 1 min read

Was this helpful?