An issue that arises in applications involving the ARCH-in-Mean (ARCH-M) model is whether or not the error variance is constant over time. A proper statistical formulation of this as a test of hypothesis presents two difficulties. First, the model does not satisfy the standard regularity...

An issue that arises in applications involving the ARCH-in-Mean (ARCH-M) model is whether or not the error variance is constant over time. A proper statistical formulation of this as a test of hypothesis presents two difficulties. First, the model does not satisfy the standard regularity...

Testing for linearity in time series models has been an active area of research [see Granger and Terasvirta (1993), Tong (1991)]. The authors consider a test for linearity against a particular regime switching model known as the smooth transition autoregressive (STAR) model.

This paper investigates whether or not the LM techniques proposed to test the null hypothesis of linearity against GARCH, bilinear (BL) and Joint GARCH-BL alternatives separately, have desirable finite sample properties. The result of a Monte carlo simulation study show that their sizes are...

This paper investigates whether or not the LM techniques proposed to test the null hypothesis of linearity against GARCH, bilinear (BL) and Joint GARCH-BL alternatives separately, have desirable finite sample properties. The result of a Monte carlo simulation study show that their sizes are...

This paper investigates whether the systematic asymmetric behaviour of the US unemployment rate can be explained by the stock market. We consider threshold models to capture the asymmetric relationship between quarterly US unemployment rate and Dow Jones Industrial Average (DJ) stock returns. We...

It is well known that, under some conditions, tests are asymptotically equivalent, and are optimal in the sense that the local power of these tests are equal to the maximum that can be achieved. However, in most practical situations, the density function f is unknown.

It is well known that, under some conditions, tests are asymptotically equivalent, and are optimal in the sense that the local power of these tests are equal to the maximum that can be achieved. However, in most practical situations, the density function f is unknown.

This paper investigates whether the systematic asymmetric behaviour of the US unemployment rate can be explained by the stock market. We consider threshold models to capture the asymmetric relationship between quarterly US unemployment rate and Dow Jones Industrial Average (DJ) stock returns. We...

This paper investigates a stock-bond portfolio s tail risks such as value-at-risk (VaR) and expected shortfall (ES) and the optimum asset allocation, and the way in which these measures have been a¤ected by the recent global financial crisis (GFC). The semiparametric method is used to estimate...