After estimating the regression parameters, an essential aspect of the analysis is to test the appropriateness of the overall model. For example, if you specify a linear regression model, but the relationship is intrinsically non-linear, then the parameter estimates (regression coefficients) and the estimated standard errors of those estimates may be significantly "off." For a review of some of the ways to evaluate the appropriateness of a model, see the following topics:

Proportion of Variance Explained

Plot of Observed vs. Predicted Values

Normal and Half-Normal Probability Plots

Variance/Covariance Matrix for Parameters

Click on the following topics for overviews of the common nonlinear regression models and nonlinear estimation procedures.