Linear regression enables the use of independent variables to forecast a dependent variable, and its simplest form involves the use of a straight line to do so, as per the equation y=mx+b, in which b is the starting point for the line (the y-intercept) and m is the slope, or steepness, of the same line.
It has wide applications in numerous fields, including electronics and computing.
For example, it can be used to gauge the price of a house on the basis of the size of its floor area, recognising that this relationship is not an exact one and is thus represented by a cloud of data points rather than a single linear relationship.
To evaluate the quality of the linear regression model, an error term is used, which is normally normalised using the R2 statistical measure.
This tool can also be employed to interpret sensor data and could, for instance, be used to create the ‘perfect’ cup of coffee.
