Hello Franciszek

this is useful website for nonparametric tests from STATA.

https://www.stata.com/features/nonparametric-methods/

Nonparametric regression techniques, including kernel and local regression, can be used in psychology research. While they may not be as commonly employed as other regression methods in psychology, they can be valuable in certain contexts where the assumptions of parametric regression models may not hold or when researchers are interested in exploring non-linear relationships.

Kernel regression, also known as kernel smoothing or kernel density estimation, is a nonparametric method that estimates the conditional expectation of a dependent variable given an independent variable. It can be particularly useful when the relationship between variables is not expected to be linear and no specific functional form is assumed. Kernel regression uses a kernel function to weight the data points around the target point, producing a smoothed estimate.

Local regression, often referred to as locally weighted scatterplot smoothing (LOWESS) or LOESS, is another nonparametric regression technique that allows for the flexible modeling of relationships. It fits a separate regression model to each data point by giving more weight to nearby points and less weight to distant points. Local regression can capture non-linear patterns and is suitable for situations where the relationship between variables may change across the range of the independent variable.

While there may be less literature specifically focused on the application of kernel regression and local regression in psychology research compared to parametric methods, it does not necessarily mean they are not useful. The choice of regression technique often depends on the specific research question, the nature of the data, and the assumptions that can be reasonably made. Parametric regression models, such as linear regression or generalized linear models, are more commonly used in psychology due to their simplicity and interpretability. However, nonparametric regression techniques can be valuable when exploring complex relationships or when assumptions of parametric models are not met.

It's worth noting that the application of nonparametric regression techniques in psychology research may be influenced by factors such as the availability of specialized software, the familiarity of researchers with these methods, and the specific research traditions within the field. Nonetheless, when appropriate, nonparametric regression approaches can offer valuable insights into relationships that may not be captured by parametric models.