The Specific Linearity Tests In Time Series

The generality of the general linearity tests has made them a popular tool for preliminary investigation and diagnostic checking in time series analysis. However, the general linearity tests do not indicate what kind of nonlinear parametric from the series should follow if they reject the null. On the other hand, the specific linearity tests reject the null in favor of the alternative model, which is well specified.This dissertation is deals with study some of the specific linearity tests and investigate the possibility to find a specific linearity test that can use to accurately determinant the alternative nonlinear model and avoiding chose alternative nonlinear model other than the true DGP series generating from, in the case of rejecting the null model. If the specific linearity test has low power against other nonlinearities, it could incorrectly accept the null of the DGP is generated from linear series.LM test is discussed as the specific linearity test. To evaluate its performance as a general linearity test, a popular general linearity test, the BDS test, is introduced for comparison. Both power and size properties of the tests are investigated using Monte Carlo simulation and Bootstrap methods.During review the previous studies, the researcher try to answer two questions, (Why researchers tend to use nonlinear time series models? What is Nonlinear Time Series Model?), as a logic introduction to this study.The researcher would like to be clear that he is not interesting here about create a new model or test, where many studies covered these points, but he is interesting about evaluate some of the exist specific linearity tests to find out which...