The usefulness of bootstrap in statistical analysis of regression models is demonstrated. Surveying earlier results, four specific problems are considered:
the computation of confidence intervals for parameters in a nonlinear regression model,
the computation of calibration sets in calibration analysis, when the standard curve is described by a nonlinear function,
the estimation of the covariance matrix of the parameter estimates for an incomplete analysis of variance model, in the presence of an interaction term,
the computation of confidence intervals for the value of the regression function, when a nonparametric heteroscedastic model is considered.
Theoretical properties of the proposed bootstrap procedures, as well as indications about their actual efficiency based on simulation results, are given.