Are gold price movements predictable? In his December 2013 paper entitled “Gold. The Bursting of a Bubble?”, Tim Verheyden assesses gold price predictability in two ways. First, he applies an autoregressive integrated moving average (ARIMA) model to assess the value of gold price technical analysis (whether past price behavior predicts future price behavior). Second, he examines interaction of gold price with inflation to assess whether the latter drives the former. Using monthly gold prices and U.S. Consumer Price Index data during January 2001 through September 2013, he finds that:
- The gold market exhibits weak form efficiency (technical analysis is not reliable).
- The relationship between gold price and inflation is weak, accounting for only 1.2% of the increase in gold price between January 2001 and September 2013.
- Therefore, it appears that gold price over the past decade represents a speculative bubble which is now deflating.
In summary, evidence suggests a speculative bubble as the most plausible explanation for gold’s dramatic price rise over the past decade.
Cautions regarding findings include:
- ARIMA modeling assumes linear relationships. To the extent that the relationship between past and future is non-linear, linear modeling may mislead. For example, Section 8.1 of “Chapter 8: Two Analysis Regimes” finds that “…the conventional SMA10 [10-month simple moving average] strategy is close to optimal for exploiting the non-linear power of the gap between the S&P 500 Index and its SMA10 to predict future index returns…”
- The author does not examine other possible indicators for gold price, such as real interest rate.
See also the broader “Future of the Price of Gold”.
