Adapted from James et al.,¹³ Springer, 2014

An algorithm is a step-by-step procedure that takes inputs and produces an output based on a set of instructions. The coefficients, or weighting of each input, are estimated by “learning” from data generated by observations or an experiment. Once the coefficients are estimated, the algorithm is known as a model and can be used to predict new outputs on data the model has not yet “seen.”