Parametric vs Non-parametric Model
The differences between parametric and non-parametric statistical learning models
Statistical learning models can be classified as parametric or nonparametric models. Machine learning models use statistical learning to predict unseen data.
The purpose of the statistical model is to approximate the relationship between the dependent and independent variables. The dependent variable is the center of attention of machine learning models which is needed to predict based on one/some independent variable.
So what is the difference between parametric and non-parametric statistical learning models?
Parametric models just like its name, we assumed the model based on a function using some parameters. So, to create the model we just need to find the parameter value by training using the existing dataset.
For example, the parametric model is Linear Regression, and the linear function uses parameters like intercept and slopes (see linear regression), so by training the data we calculated these parameters.
Non-parametric models do not assume the model based on a function. this model is flexible, the example of a non-parametric model is the K-Nearest Neighbor.
The advantage of the parametric model compared to the non-parametric model is its flexibility. If the relationship between the dependent and independent variables is not based on a function, the non-parametric model can fit in nicely. Otherwise, the advantage of a parametric model is when the relationship is based on a function, this model will nicely predict the future unseen data, using a non-parametric model will tend to overfit the model to the training data compared to the parametric model.
So which one is the best? it depends, we need to do some exploratory data analysis, trial and error using some models to the training and test data to find the best model.
