Abstract:We investigate statistical properties of daily international market indices of seven countries, and high-frequency $S&P500$ and KOSDAQ data, by using the detrended fluctuation method and the surrogate test. We have found that the returns of international stock market indices of seven countries follow a universal power-law distribution with an exponent of $\zeta \approx 3$, while the Korean stock market follows an exponential distribution with an exponent of $\beta \approx 0.7$. The Hurst exponent analysis of the original return, and its magnitude and sign series, reveal that the long-term-memory property, which is absent in the returns and sign series, exists in the magnitude time series with $0.7 \leq H \leq 0.8$. The surrogate test shows that the magnitude time series reflects the non-linearity of the return series, which helps to reveal that the KOSDAQ index, one of the emerging markets, shows higher volatility than a mature market such as the {S&P} 500 index.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is to explore the statistical characteristics of international stock market indices (including daily market indices of seven countries and high - frequency S&P500 and KOSDAQ data), especially the return distribution characteristics of these markets, their long - term - memory property and non - linearity. Specifically:
1. **Return Distribution Characteristics**: The study found that, except for the South Korean stock market, the returns of the other seven international stock markets follow a general power - law distribution, with an exponent of approximately \(\zeta \approx 3\). The returns of the South Korean stock market follow an exponential distribution, with an exponent of approximately \(\beta \approx 0.7\).
2. **Long - term Memory**: By analyzing the Hurst exponent of the original returns, magnitude series and sign series, the study found that long - term memory mainly exists in the magnitude series, and its Hurst exponent \(H\) ranges from 0.7 to 0.8. In the returns and sign series, this long - term memory is not obvious, showing the characteristics of a random walk, that is, \(H \approx 0.5\).
3. **Non - linear Characteristics**: By comparing the original time series and surrogate data, the study found that the magnitude series reflects the non - linear characteristics of the original returns, while the sign series shows linear characteristics. In addition, the study also found that the KOSDAQ index, as an emerging market, shows higher volatility than mature markets such as S&P500, which may be caused by the abnormal generation of market bubbles.
In summary, through detailed statistical analysis methods, this paper reveals the differences between different markets, especially their different performances in terms of return distribution, long - term memory and non - linear characteristics. These findings are helpful for better understanding the complexity of the global financial market, especially in distinguishing between mature and emerging markets.