GMM ESTIMATION AND SHAPIRO-FRANCIA NORMALITY TEST: A CASE STUDY OF DEVELOPING COUNTRIES

The study of relationships between economic and financial system indicators in developing countries became contemporary issue taking into account the influence of the last crisis. This paper aims to analyze the correlation of number of explanatory variables with portfolio investments in 38 emerging markets in the period from 2005 to 2009. As a empirical methodology of this research serves Arellano-Bond and Arellano/Bover-Blundell/Bond estimations and Shapiro-Francia normality test. We identify that the capital market indicators, in particular, the stock market capitalization, stock market total value traded to GDP are considered the main positive and statistically significant variables. Since the outbreak of the crisis, investors start to pay their attention to almost every indicator affecting the flows of portfolio investments. Almost all banking and economic indicators (except some business environment indicators, bank non-performing loans to gross loans (%), lending interest rate (%), banks Z-score and official exchange rate) are highly associated with portfolio investments.