Effects of skewness and kurtosis on model selection criteria

Abstract

We consider the behavior of model selection criteria in AR models where the error terms are not normal by varying skewness and kurtosis. The probability of estimating the true lag order for varying degrees of freedom (k) is the interest. For both small and large samples skewness does not effect the performance of criteria under consideration. On the other hand, kurtosis does effect some of the criteria considerably. In large samples and for large values of k the usual asymptotic theory results for normal models are confirmed. Moreover, we showed that for small sample sizes performance of some newly introduced criteria which were not considered in Monte Carlo studies before are better.

A Multivariate Approach for Identification of Optimal Locations with in Ethiopia’s Wheat Market to Tackle Soaring Inflation on Food Price

“…Means for these 6 markets there is no second unit root. For Addis Ababa and Nazret the null of unit root around zero under 1 Φ and unit root around drift under t α is rejected at 5% level but not at 10% level 10 . However given the data is having zero mean it is known fact that t α and 1 Φ have weak power compared to df t .…”

“…So we have to reject the Mekelle's test result, too. When the 1 Φ version of the F test is used the null of unit root around zero is not rejected even at 10% level for all but Addis Ababa. For Addis Ababa the null is rejected at 1% level which could imply that there is either unit root around drift, drift with stationery series or unit root around drift.…”

“…So the only logical conclusion is that the data is following unit root around zero. But under such assumption both 1 Φ and t α have weak power so it have to be proven by more power full ADF test of df t . Assuming there is drift in the level data of Gonder at 10% and Nazret at 5% level are having stationary series.…”

“…In short run supply shocks and inflation inertia followed by money supply growth are strong determinants of inflation. It is hard to swallow the above result when most of the inflation is observed in staple grains which are none tradable for Ethiopia 1 and when oil price is increasingly subsidized by state with every increase in international price of oil. Moreover a study by Ulimwengu et al (2009) did show that domestic maize price are not cointegrated with international maize price, which support the view that Ethiopian food inflation, which is the major cause of the over all inflation, is not caused by Global or regional factors but by domestic factors.…”

Three essays on data contaminants, outliers and macroeconomic time series

“…An asterisks indicates statistical significance at the 1-percent level. Zaman (1998) establish that excess kurtosis, which can be interpreted as non-contamination-based innovation outliers, affects various model selection criteria but only marginally. Le, Raftery, and Martin (1996) argue that additive outliers create downward bias in automatic selection criteria.…”