LINEAR REGRESSION METHOD
Reminder on the linear regression method
We use the linear regression method only when the absolute uncertainty on x is negligible (Δx ≈ 0) and the absolute uncertainty on y is constant .
This method consists in drawing the best possible straight line of equation y=a+bx which passes as close as possible to all the experimental points, as illustrated in figure 1.
b represents the slope of the regression line while is the ordinate at the origin.
Knowing
that we have n measurements performed in practical work, then
the values of a and b
are given by the formulas:
The uncertainties associated with each parameter are given by:
If the linear regression line passes through the origin, we have the equation y = ax of the line ,
in this case a = 0 and we obtain:
