An analysis of short - term and long - term relationships between Borsa iIstanbul 100 & Tourism Index, and some developed stock market indexes

A remarkable development over the last 20 years has been the decline in the number of restrictive practices on capital flow occurring as a result of financial liberalization. The development and liberalization of financial markets, and the increase in interaction among sub-markets have made share prices extremely vulnerable to macroeconomic factors and political events. Any development or uncertainty experienced by a given market can now affect other markets easily in a very short time. Although the relationship between stock prices and macroeconomic variables has been studied since the 1970's, the number of studies analysing the relationship between domestic stock markets and foreign ones has been limited. Furthermore, the tourism sector, a very important factor for domestic economies not only due to its contribution to domestic income but also the inflow of receipts it provides, is one of the fastest-developing sectors in world economy. As a result of foreign expansion and liberalization policies that began by the early 1980's, tourism, like international trade in goods and services, has become an important issue, and investment in tourism has been promoted by governments. In line with these developments, Borsa İstanbul National 100 Index and Borsa İstanbul Tourism Index have been studied, and the relationship between these two indexes and indexes of seven developed countries, America (S&P 500), England (FTSE 100), Germany (DAX), France (CAC 40), Italy (FTSEMIB), Canada (S&PTSX) and Japan (NIKKEI 225) have been examined. Data pertaining to a period of 206 months (between 31.12.1997 and 30.01.2015) has been analysed. Augmented Dickey Fuller (ADF) and Philips Peron (PP) unit root tests have first been applied to the series the logarithms of which were taken. Upon the stabilization of the series, long-term relationships for cointegrated variables have been produced by means of Vector Error Correction Model. For non-cointegrated variables, short-term relationships have been determined through Vector Autoregressive Model. As a result of the analyses made, no long-term causal relations have been found between indexes of developed countries and Borsa İstanbul 100 (BİST 100) and Borsa İstanbul Tourism (XTRZM) indexes. Moreover, it has been found that whereas in short term Canadian (S&PTSX) index is a cause of only Borsa İstanbul 100 (BİST 100) index, French (CAC 40) and Italian (FTSEMIB) indexes are causes of both Borsa İstanbul Tourism index and Borsa İstanbul 100 index. Lastly, the cointegration analyses of variables in pairs have provided supporting evidence for preliminary findings. Consequently, it can be said that investors may get an idea of BİST 100 index through Borsa İstanbul Tourism index, and that people who have invested in BİST 100 index can diversify their portfolio by investing in the seven aforementioned stock markets in order to reduce risks and increase returns. Key words: BİST 100, XTRZM, cointegration, VAR, VECM