Testing the Empirical Validity of the Feldstein-Horioka Puzzle for Gambia Using the Gregory Hansen Cointegration Test and ADRL

The paper tests the empirical validity between investment and saving for the Gambia by employing co-integration test with structural break and regimes shifts. The data was obtained from WDI. The ARDL model with dummies and interaction variables is used to determine the short run, long run and effect of structural breaks in the investment saving relationship to test the empirical validity for The Gambia. The results showed the existence of break points at 1983 and 2005 and no co-integration at the break points. The absolute values of ADF, , are all less than 5% critical values, therefore we failed to reject the null hypothesis of no co-integration at the break points. ADRL with break dummies shows that the long run coefficient for saving (saving retention coefficient) is negative and significant but openness is positive and not significant. The break dummy is negative and highly significant which shows that the break is important in modeling the relationship between investments and saving. The interaction term of investment is positive and highly significant at 5%, but the interaction term for saving is positive and not significant. In the Short run, the coefficient of saving to output is positive (β = .5493598) and significant at 1% and trade is also positive and significant at 5%. ADRL without break dummies shows that the long run coefficient for saving and openness are positive and not significant and in the short run they are positive and significant.

See more details : - www.globalpresshub.com