Temple (1998) observes that outliers that arise from measurement error, omitted variables or
parameter heterogeneity can bias the results of growth models. To address the issue of parameter
heterogeneity and influential outliers, the transition to steady state models are re-estimated using
the robust regression technique which gives minimum weight to outlying observations. The results
are reported in Table 7. The estimates are not significantly different to the OLS or GMM results
reported in Tables 5 and 6, suggesting that the results are not being driven by outliers. The
coefficients on the stock market variables are statistically significant indicating the importance of
the stock market for economic growth.