again largely similar to the findings of the full sample. Platinum volatility is also
impacted by stock market volatility in the first half of the sample, although this was not
observed in the full sample analysis. Interestingly silver volatility, is largely unaffected
by any of the macroeconomic variables for the full and first subperiod, although there is
some evidence of effects from the term structure variable upon silver returns in the
second half. This would be consistent with expectations of future silver demand based
upon business cycle effects..
Overall, analysis in the second half of the sample, however, paints a different picture to
the earlier findings. In summary, our models lose their explanatory power in the latter
part of the sample as none of the macroeconomic variables figures significantly. This
could indicate changes in the dynamics of these markets and the fundamental variables
that affect them. It is also possible that this finding is due to the great price increase in the
precious metals markets, which occurred during the second half. If the momentum effects
dominate and are largely responsible for volatility in these markets, the statistical analysis
may not capture the linkage with more fundamental macroeconomic variables in a
subperiod analysis. It would be of interest to investigate this issue further in future work
to uncover the underlying market dynamics.
5. Concluding Remarks
The key objective of this paper was to investigate and present the key macroeconomic
factors that impact the price returns of precious metals markets. The markets investigated
were gold, silver, platinum and palladium, whereas the macroeconomic factors
14