Page 133 - Contributed Paper Session (CPS) - Volume 5
P. 133

CPS1144 Adeniji Nike Abosede et al.
            17889 unit increase will lead to oil export. Since the dynamic mode was fixed
            and  a  positive  relationship  were  examined,  we  go  ahead  and  studied  the
            forecast error variance decomposition i.e. the variation explained due to shock
            on each other using eight forecast time horizon. Tables 1, displace the forecast
            error variance decomposition at both aggregate and disaggregate level.

             Period      Non – oil export          Oil export           Total export
                1           18.39643                8.868136             9.996062
                2           18.42322                14.85511             13.61430
                3           18.44418                15.18785             13.62520
                4           18.47710                15.27060             13.63614
                5           18.52831                15.27970             13.63624
                6           18.60795                15.28121             13.63628
                7           18.73163                15.28140             13.63629
                8            18.9234                15.28143             13.63730
                Table  1:  Forecast  error  variance  of  naira/dollar  exchange  rate  on  total
            export, non- oil, and oil export from 1980  -2010 using eight forecast time
            horizons

            4.  Discussion and Conclusion
                In other not to have a spurious regression results from a time series data,
            we subjected each variable to unit root test through the use of ADF test under
            two different assumptions that the data generating mechanisms followed. We
            address the forecast error variance decomposition on the impact of foreign
            exchange rate volatility on Nigeria export growth using annual time series data
            for period 1980 to 2010 using time series technique which encompasses unit
            root  test,  vector  auto  regression  to  look  at  the  dynamic  relationship  and
            variance decomposition to study the changes in variation of variable to one
            another. The error derived from eq(7) showed that it explained all the forecast
            error variance of non – oil, total oil and oil export at all eight forecast time
            horizon, indicating that non – oil, oil, and total export can be treated as pure
            endogenous variables. In this sense, past and current values of naira/dollar
            exchange rate can  be  used  in  predicting  each of  the  dependent  variables.
            Naira/dollar exchange rate had slight effect on export at both aggregate and
            disaggregates  level.  For  improvement  on  Nigeria  export  growth,  a  well
            appreciable  mechanism  should  be  adopted  either  through  devaluation  or
            currency  restructuring  which  will  aid  policy  decision  makers,  investors  and
            other  business  attributes  to  have  efficient  productivity  on  Nigeria  export
            growth.





                                                               122 | I S I   W S C   2 0 1 9
   128   129   130   131   132   133   134   135   136   137   138