Abstract:
The Ghanaian economy is dependent to a high degree on primary production, in agriculture and
mining for exports. The exports of cocoa, gold and timber traditionally account for the greatest
bulk of merchandise exports. The production of cocoa beans is thought to employ more than 1.5
million Ghanaians. Cocoa is still the single-most important commodity to the country's economy.
For this study econometric methods were used to explore the patterns of domestic production
volatility of cocoa under consideration in Ghana from October 2000 to September 2015 major
crop seasons by developing GARCH model. From the results it was observed that daily cocoa
purchases show a higher purchase follow lower purchase. In particular, high increases of
purchases are observed in the October 2011. Crop purchases for 2014/2015 were the lowest
over the past six main cocoa seasons. The empirical result shows the average purchase per
tonne for cocoa to be 18,896.2 with standard deviation of 17,852.1. The series were positive
Skewness (2.3) and longer tails. Excess kurtosis coefficients 12.3 indicated that the distribution
of purchase series for cocoa possess leptokurtic characteristics. The Jarque-Bera test statistic
indicates that the purchases series is non-normality. Further, the computed ADF test - statistic
(-20.80127) was seen to be significant, hence the return of cocoa purchases series doesn't
have a unit root problem. The ARCH and GARCH coefficients (0.53 and 0.15) are statistically
significant which indicate that shocks to volatility have a persistent effect on the conditional
variance. The conditional standard deviation shows periods of high volatility.
