Kaushik Matia,Yosef Ashkenazy,H. Eugene Stanley
Abstract:We analyze daily prices of 29 commodities and 2449 stocks, each over a period of $\approx 15$ years. We find that the price fluctuations for commodities have a significantly broader multifractal spectrum than for stocks. We also propose that multifractal properties of both stocks and commodities can be attributed mainly to the broad probability distribution of price fluctuations and secondarily to their temporal organization. Furthermore, we propose that, for commodities, stronger higher order correlations in price fluctuations result in broader multifractal spectra.
What problem does this paper attempt to address?
This paper aims to explore the multifractal characteristics of stock and commodity price fluctuations. Specifically, the author analyzed the daily price data of 29 commodities and 2,449 stocks over approximately 15 years and found that the price fluctuations of commodities have a wider multifractal spectrum than stocks. The paper further proposes that these multifractal characteristics are mainly attributed to the wider probability distribution of price fluctuations, and secondly to their temporal organizational structure. For commodities, the higher - order correlations in price fluctuations lead to a wider multifractal spectrum.
### Main research questions:
1. **Compare the price fluctuation characteristics of stocks and commodities**: By analyzing the price fluctuations of 29 commodities and 2,449 stocks, explore the differences in multifractal characteristics between the two.
2. **Sources of multifractal characteristics**: Investigate whether multifractal characteristics are mainly determined by the probability distribution of price fluctuations or whether the temporal organizational structure also plays an important role.
3. **Differences between the commodity market and the stock market**: Pay special attention to how the special properties of the commodity market, such as storage and transportation requirements, and the response speed to demand changes, affect its multifractal characteristics.
### Research methods:
- **Data processing**: Define the standardized price fluctuation (return value) \( g(t)\equiv\frac{\ln S(t + \Delta t)-\ln S(t)}{\sigma}\), where \(\Delta t = 1\) day, \(S(t)\) is the price, and \(\sigma\) is the standard deviation of \(\ln S(t+\Delta t)-\ln S(t)\).
- **Multifractal detrended fluctuation analysis (MF - DFA)**: Use the MF - DFA method to study the multifractal characteristics of stocks and commodities, including calculating the profile of the time series, local trends, mean - square fluctuations, etc.
- **Randomization treatment**: Destroy the time correlation by randomly exchanging data points in the time series and observe the changes in the multifractal characteristics after randomization.
### Main findings:
- **Commodities have a wider multifractal spectrum**: The price fluctuations of commodities have a wider multifractal spectrum than stocks.
- **Sources of multifractal characteristics**: Multifractal characteristics are mainly determined by the probability distribution of price fluctuations and secondly influenced by the temporal organizational structure.
- **Special properties of the commodity market**: The special properties of the commodity market, especially the slow response to demand changes, lead to a wider multifractal spectrum.
Through these analyses, the paper reveals the significant differences in multifractal characteristics between stock and commodity price fluctuations and proposes possible explanatory mechanisms.