Spectral Analysis of Overseas Tourists’ Expenditures in the United Kingdom

Abstract

This article introduces univariate and bivariate spectral analysis as methods of modeling overseas tourists’ expenditures in the United Kingdom. In general, these two methods detect cyclical behavior within and between time-series data, respectively. The univariate spectrum refutes any long-term “business cycle”-type oscillations in the expenditure data. Only a significant annual cycle is revealed. The univariate spectrum is shown to assist greatly in the difficult problem of parameter determination in the more conventional autoregressive integrated moving average (ARIMA) class of time-series models. The bivariate spectrum reveals significant lead relationships between cycles in pound sterling exchange rates against the U.S. dollar and French franc and cycles in tourists’ expenditures. Such relationships would be difficult to detect via more commonly used time-series procedures.