1Reading 2 Gujarati Chapter 9 When using log-linear or log-log models the elasticity coefficient is constant but the slope coefficient is variable. The LIV models however have a constant slope and the elasticity coefficient is variable. There are five major types of nonlinear-in-variable but linear -in-parameter models. They are: log-linear, log-lin, lin-log, reciprocal model, and polynominal models. We can combine features of these models to come up with a multiple regression model that has a dependent variable in log form and some of the X variables in log for, with some in linear form. Reading 3 Gujarati Chapter 14 The autocorrelations can be biased as with variances and errors. They can be unbiased when used with OLS estimators although they are not efficient. This means that standard t and z tests can be misleading at a large scale. By using the runs test, plotting of residuals and the Durbin-Watson d test we can determine if there is autocorrelation. Once it has been found it is important to
This is the end of the preview.
access the rest of the document.