Emerging stock market co-movements in South Asia: wavelet approach

The purpose of this paper is to study the co-movement and market integration dynamics of the emerging/frontier stock markets in South Asia (India, Pakistan and Sri Lanka) with a portfolio management perspective.

Scholars in the past have documented the limitation of standard econometric techniques such as co-integration analysis to capture this phenomenon. The other econometric technique widely used in integration and comovement literature is dynamic conditional correlation-generalized autoregressive conditional heteroskedasticity. This method captivates the time-varying correlations, although frequency information is absent. The wavelet-based analysis decomposes the time-series data in a time-frequency domain, which is largely useful to fund managers and policy makers. This study examines the regional integration in selected South Asian markets using wavelet analysis.

The results suggest some degree of market integration, however weak as compared to regional integrations in developed markets. Pakistan and India were found to be the potential leaders at varying time scales in the region. Weaker co-movement phenomena may offer ample arbitrage opportunities to investors in this region. In addition, the authors also find that the structure of correlation changes after some of the major macroeconomic events.

This study is among the first to examine co-movement and integration of stock returns in a time-frequency domain for South Asia. In addition, the authors also highlight weak integration in these markets, which may be beneficial for portfolio diversification.