An overview of global gold market and gold price forecasting. 2010.
This paper does two things. Firstly it gives an overview of some aspects of the gold market and finds the following. The average ratio of the gold to oil is 11 barrels of oil per ounce of gold. The show a negative correlation between gold and inflation between 1968 and 2009 of -9% which the authors posit as proof that there is no significant relationship between the two variables.
They also model and attempt to predict the real gold price as a process in which the real price of gold follows a price path based on a trend stationary process, with jumps or dips from the long run trend. These are shown to be a significant part of the movements of the gold price over time. From this model the authors predict that the real price of gold would remain above trend until 2014 at which time it would revert to its trend stationary process.
Data: Gold supply and demand data from the World Gold Council. Monthly gold prices.
Methodology: Trend Stationary Process analysis with jump and dip diffusion models. Unit roots.
Citation: Shafiee, Shahriar, and Erkan Topal. “An overview of global gold market and gold price forecasting.” Resources Policy 35.3 (2010): 178-189.
Abstract: The global gold market has recently attracted a lot of attention and the price of gold is relatively higher than its historical trend. For mining companies to mitigate risk and uncertainty in gold price fluctuations, make hedging, future investment and evaluation decisions, depend on forecasting future price trends. The first section of this paper reviews the world gold market and the historical trend of gold prices from January 1968 to December 2008. This is followed by an investigation into the relationship between gold price and other key influencing variables, such as oil price and global inflation over the last 40 years. The second section applies a modified econometric version of the long-term trend reverting jump and dip diffusion model for forecasting natural-resource commodity prices. This method addresses the deficiencies of previous models, such as jumps and dips as parameters and unit root test for long-term trends. The model proposes that historical data of mineral commodities have three terms to demonstrate fluctuation of prices: a long-term trend reversion component, a diffusion component and a jump or dip component. The model calculates each term individually to estimate future prices of mineral commodities. The study validates the model and estimates the gold price for the next 10 years, based on monthly historical data of nominal gold price.