Impact of Drought on Commodity Market Forecasting
|Author:||Bc. Kristian Racocha|
|Year:||2020 - summer|
|Leaders:|| Marek Hauzr
|Work type:|| Bachelors
|Awards and prizes:|
|Abstract:||This thesis analyzes the relationship of drought (SPEI-01) and commodity
prices. Cyclical properties of both components are extracted and by utilizing the spectral factor models proposed by Bandi et al. (2019) the impact of
SPEI-01 on various commodity prices across a range of frequencies is examined.
The analysis reveals that most of the commodities are impacted by SPEI-01
when considering cycles of lengths 2 − 4 months and 32 − 64 months. For
frequencies, where a signifcant relationship is found, the commodity prices
are forecasted using the ARIMA model with and without the inclusion of the
drought variable. The results suggest that the inclusion of drought is benefcial
for a majority of the models, however, in a considerable number of occasions
the models exhibited a diminished performance.