Moving Block Bootstrap Method for Determining Confidence Intervals for a Change Point in Time Series in the Presence of Conditional Heteroscedasticity

This paper seeks to use the Moving Block Bootstrap method to determine the confidence intervals for a change point in conditional variance function of data exhibiting conditional heteroscedasticity and heterogeneity. Confidence intervals for a change point normally provide or give a range within which the true change point location is likely to lie. This is usually based on a specified confidence level. This helps to in turn determine whether the change point is statistically significant especially after determining the critical values for the distribution of the change point test statistic. Confidence intervals are also called interval estimates as opposed to point estimates which provide a single estimate for a parameter.