Nonlinear Regression and Modeling

Many functional relationships do not lend themselves to Unearization the user has to choose either option 3 or 4 in Section 2.3 to continue. 

When confronted with multidimensional data it is easy to plug the figures into a statistical package and have nice tables printed that purportedly accurately analyze and represent the underlying factors. Have the following questions been asked  

Does the model conform to the problem Is the number of factors meaningful  

The point made here is that nearly any model can be forced to fit the data the more factors (coefficients) and the higher the order of the independent variable(s) (x, x, x, etc.) the better the chance of obtaining near-zero residuals and a perfect fit. Does this make sense, statistically, or chemically  

Visualizing Data, the reader may have guessed from previous sections that graphical display contributes much toward understanding the data and the statistical analysis. This notion is correct, and graphics become more important as the dimensionality of the data rises, especially to three and more dimensions. Bear in mind that  

---