Fortschritte in der angewandten Wissenschaftsforschung Offener Zugang

Abstrakt

Neural network applications in the forecasting of GDP of Nigeria as a function of key stock market indicators

Olaniyi S Maliki, Ibina Emmanuel and Eze E. Obinwanne

Stock market provides the bridge through which the savings of surplus units may be transformed into medium and long-term investments in the deficits units. It is reputed to perform critical functions, which promote economic growth and prospects of the economy. Empirical evidence linking stock market development to economic growth has been inconclusive even though the balance of evidence is in favour of a positive relationship between stock market development and economic growth. This paper explores the relationship between stock market and economic growth in Nigeria and attempts to construct its validity or otherwise, using quarterly data from 1990:Q1 to 2009:Q4 for Nigeria by employing artificial neural network as a predictive tool and comparing with the statistical method of multiple regression analysis. Here we take the GDP representing economic growth for this period, as a function of key stock market indicators including Market Capitalization (MC), Number of Deals (ND), All-Share Value Index (ASVI), Total Value of Shares Traded (TVST), and Inflation Rate (IR). From the multivariate statistical analysis, it becomes evident that the independent variables of our model are all highly correlated with the dependent variable of GDP, excepting the inflation rate having a negative correlation value of approximately 0.30. However, the value of the goodness of fit (R2) is given as 0.85. For comparison of efficiency between ANN and regression analysis we computed five performance measures and it was discovered that ANN outperforms regression analysis significantly, and is thus better suited to stock market prediction