Modelling a latent daily Tourism Financial Conditions Index

Abstract

The paper uses daily data on financial stock index returns, tourism stock sub-index returns, exchange rate returns and interest rate differences from 1 June 2001–28 February 2014 for Taiwan to construct a novel latent daily tourism financial indicator, namely the Tourism Financial Conditions Index (TFCI). The TFCI is an adaptation and extension of the widely-used Monetary Conditions Index (MCI) and Financial Conditions Index (FCI) to tourism stock data. However, the method of calculation of the daily TFCI is different from existing methods of constructing the MCI and FCI in that the weights are estimated empirically. Alternative versions of the TFCI are constructed, depending on the appropriate model and method of estimation, namely Ordinary Least Squares (OLS) or Quasi-Maximum Likelihood Estimation (QMLE) of alternative conditional volatility models. Three univariate conditional volatility models are considered, namely GARCH, GJR and EGARCH, in an attempt to capture the inherent volatility in the daily tourism stock index returns. The empirical findings show that TFCI is estimated quite accurately using the estimated conditional mean of the tourism stock index returns, especially when conditional volatility is incorporated in the overall specification. The new daily TFCI is straightforward to use and interpret, and provides interesting insights in predicting the current economic and financial environment for tourism stock index returns, especially as it is based on straightforward calculations and interpretations of publicly available information.