Nonparametric regression

Nonparametric regression is a type of regression (a way of predicting something from data) where the parameters that are used to describe the relationship between the data points do not have an fixed number. This is in contrast to a parametric regression, where the parameters have a fixed number. With nonparametric regression, the parameters are determined from the data we have. This means that the computer looks at the data we have and figures out what parameters fit the data best, without us telling it that there needs to be a certain number of parameters.