The main aim of this chapter is to provide an econometric analysis focused on investigating the consequences of financial contagion between certain developed capital markets, such as USA, France, UK, and Germany in terms of global financial crisis. In the recent past, the impact of international transmission mechanisms significantly affected the investment behavior due to the propagation of financial shocks. More specifically, the risk of financial contagion highlights the vulnerability of traditional assumptions based on efficiency and rationality considering the global implications of resource allocation performance and international portfolio diversification.
TopData Series And Empirical Analysis
The empirical analysis is based on daily stock returns of selected stock markets major indices for the period January 2007 until March 2013 with the exception of legal holidays or other events when selected capital markets haven’t performed financial transactions.
The Augmented Dickey-Fuller (ADF) test is used in order to determine the non-stationarity or the integration order of a financial time series. Practically, the ADF diagnostic test investigates the existence of unit roots and includes the following categories: unit root with a constant and a trend, unit root with a constant, but without a time trend, and finally unit root without constant and temporal trend, based on the following regression model:
where p represents the number of lags for which it was investigated whether fulfilling the condition that residuals are white noise, c is a constant, t is the indicator for time trend and Δ is the symbol for differencing. The stochastic trend cannot be predicted due to the time dependence of residual’s variance. Moreover, in case that the coefficients to be estimated β and δ have the null value then the analyzed financial time series is characterized by a stochastic trend. The null hypothesis, i.e. the time series has a unit root is rejected if t-statistics is lower than the critical value.
The BDS test (Brock, Dechert and Scheinkman, 1987) was computed in order to determine whether the residuals are independent and identically distributed, especially in the case of a nonlinear system. In this respect BDS test is an econometric tool for detecting serial dependence in financial time series. Technically, the BDS diagnostic test is based on the null hypothesis of independent and identically distributed (i.i.d) time series.
The Granger causality test is mainly considering the possible relationships between two or more time series. Basically, it is a statistical hypothesis test that highlights the possible influence of a time series data in forecasting another series. Primary, causality highlights a relationship between cause and effect.