Abstract:
|
Temporally aggregated data with calendrical periods, e.g., weekly, monthly, quarterly, or annually, have been widely used because the aggregation technique is simple and convenient for summarizing long sequential measurements and reducing their data length. However, it is inevitable that temporal aggregation causes a significant loss of information. Thus, when disaggregating the totals, restoring the original information remains a considerable challenge in time series analysis.
In this research, we will propose an interval estimation method to trace an unknown disaggregate time series within certain bandwidths. First, we will consider the two model-based disaggregation methods, called the generalized least squares (GLS) disaggregation and the ARIMA disaggregation. Then, we will develop iterative steps to construct AR-sieve bootstrap prediction intervals for the model-based temporal disaggregation. As an illustration, we will analyze the quarterly total balances of the U.S. international trade in goods and services between the 1st quarter of 1992 and the 4th quarter of 2020.
|