ABSTRACT
The non-structural econometric modelling approach is particularly suited to the analysis of stock market relationships given the theoretical difficulties involved in capturing all the variables that affect equity market price. Granger-causality analysis pinpoints the direction of causation while the forecast variance decomposition and impulse response analyses, which are readily available within vector autoregression (VAR), measure the duration and speed of interaction between the equity markets. This chapter discusses VAR in terms of its concept, model, requirements, strengths and weaknesses, and its use in equity market integration studies. It then explains the unit root test and lag test used in this study. The VAR which serves as the basis for all the econometric techniques exemplifies the non-structural modelling approach. The chapter discusses in detail the different time series techniques. These techniques are cointegration, error correction model, Granger-causality, forecast variance decomposition and impulse response analyses.
