Identifying influential observations in nonlinear regression
Not all observations are of equal importance for the results of a regression analysis. Observations that highly influence the inference are called influential observations and the detection of such observations is of great importance. In this presentation I will focus on the detection of observations with high influence on the parameter estimates in a nonlinear regression model. The use of influence measures for detection of influential observations will be illustrated with numerical examples. Moreover, the approach for constructing the measures will be presented together with their explicit expressions.