Guest lecture by Radim Navrátil on February 11th

Doctoral candidate Radim Navrátil from the Charles University in Prague will hold a presentation on rank tests in heteroscedastic linear model with nuisance parameters. The lecture will take place on Tuesday, 11th of February at 10:15 in the seminar room 469, Publicum 4th floor.

Abstract:

Homoscedasticity is often tacitly assumed in the analysis of linear models, both classical and robust. To avoid a negative consequence of ignored heteroscedasticity, we should either analyze its possibility before starting an inference on the parameters of the model, or look for an approach invariant to heteroscedasticity, if there is one.

In the linear regression model with heteroscedastic errors, we propose nonparametric tests for regression under nuisance heteroscedasticity, and tests for heteroscedasticity under nuisance regression. Both types of tests are based on suitable ancillary statistics for the nuisance parameters; hence they avoid their estimation, in contradistinction to tests proposed in the literature. A simulation study, as well as an application of tests to real data, illustrate their good performance

Guest lecture by Radim Navrátil on February 11th

Doctoral candidate Radim Navrátil from the Charles University in Prague will hold a presentation on rank tests in heteroscedastic linear model with nuisance parameters. The lecture will take place on Tuesday, 11th of February at 10:15 in the seminar room 469, Publicum 4th floor.

Abstract:

Homoscedasticity is often tacitly assumed in the analysis of linear models, both classical and robust. To avoid a negative consequence of ignored heteroscedasticity, we should either analyze its possibility before starting an inference on the parameters of the model, or look for an approach invariant to heteroscedasticity, if there is one.

In the linear regression model with heteroscedastic errors, we propose nonparametric tests for regression under nuisance heteroscedasticity, and tests for heteroscedasticity under nuisance regression. Both types of tests are based on suitable ancillary statistics for the nuisance parameters; hence they avoid their estimation, in contradistinction to tests proposed in the literature. A simulation study, as well as an application of tests to real data, illustrate their good performance