What Non-parametric regression is
Non-parametric regression is a type of regression analysis used when the data is non-linear or when there is no clear relationship between the independent and dependent variables. It is used to estimate the relationship between two or more variables without making any assumptions about the underlying distribution of the data.
Steps for Non-parametric regression:
- Define the independent and dependent variables.
- Collect data for the variables.
- Choose a non-parametric regression model.
- Fit the model to the data.
- Evaluate the model.
- Interpret the results.
Examples
- Estimating the median of a variable from a dataset when the underlying distribution is unknown.
- Estimating the relationship between two variables when the functional form of the relationship is unknown.
- Estimating the probability of a certain event occurring based on observed data when the functional form of the relationship is unknown.
- Estimating the expected value of a variable from a dataset when the underlying distribution is unknown.