In Part I we introduced the classical linear regression model with all its assumptions. In Part II we examined in detail the consequences that ensue when one or more of the assumptions are not satisfied and what can be done about them. In Part III we study some selected but commonly encountered econometric techniques. In particular, we discuss these topics: (1) nonlinear-in-the-parameter regression models, (2) qualitative response regression models, (3) panel data regression models, and (4) dynamic econometric models.
In Chapter 14, we consider models that are intrinsically nonlinear in the parameters. With the ready availability of software packages, it is no longer a big challenge to estimate such models. Although the underlying mathematics may elude some readers, the basic ideas of nonlinear-in-the-parameter regression models can be explained intuitively. With suitable examples, this chapter shows how such models are estimated and interpreted.
In Chapter 15, we consider regression models in which the dependent variable is qualitative in nature. This chapter therefore complements Chapter 9, where we discussed models in which the explanatory variables were qualitative in nature. The basic thrust of this chapter is on developing models in which the regressand is of the yes or no type. Since OLS poses several problems in estimating such models, several alternatives have been developed. In this chapter we consider two such alternatives, namely, the logit model and the probit model. This chapter also discusses several variants of the qualitative response models, such as the Tobit model and the Poisson regression model. Several extensions of the qualitative response models are also briefly discussed, such as the ordered probit, ordered logit, and multinomial logit.
In Chapter 16 we discuss panel data regression models. Such models combine time series and cross-section observations. Although by combining such observations we increase the sample size, panel data regression models pose several estimation challenges. In this chapter we discuss only the essentials of such models and guide the reader to the appropriate resources for further study.
In Chapter 17, we consider regression models that include current as well as past, or lagged, values of the explanatory variables in addition to models that include lagged value(s) of the dependent variable as one of the explanatory variables. These models are called, respectively, the distributed lag and autoregressive models. Although such models are extremely useful in empirical econometrics, they pose some special estimating problems because they violate one or more assumptions of the classical regression model. We consider these special problems in the context of the Koyck, the adaptive-expectations (AE), and the partial-adjustment models. We also note the criticism leveled against the AE model by the advocates of the so-called rational expectations (RE) school.
Was this article helpful?
Learning About The Rules Of The Rich And Wealthy Can Have Amazing Benefits For Your Life And Success. Discover the hidden rules and beat the rich at their own game. The general population has a love / hate kinship with riches. They resent those who have it, but spend their total lives attempting to get it for themselves. The reason an immense majority of individuals never accumulate a substantial savings is because they don't comprehend the nature of money or how it works.