ABSTRACT
The techniques developed in Chapter 11 estimate using linear (or iteratively linear in the case of two-stage least squares and the generalized method of moments) procedures. The linearity was particularly valuable before the widespread availability of computers and the development of more complex mathematical algorithms. However, the innovations in computer technology coupled with the development of statistical and econometric software have liberated our estimation efforts from these historical techniques. In this chapter we briefly develop three techniques that have no closed-form (or simple linear) solution: nonlinear least squares and maximum likelihood, applied Bayesian estimators, and least absolute deviation estimators.
