Mean reversion in stock prices of seven Asian stock markets: Unit root test and stationary test with Fourier functions

- Analyze mean reversion of seven Asian stock markets.
- Use both LM Fourier unit root test and a stationary test with a Fourier function.
- The LM Fourier unit root test shows that stock prices are all mean-reverting.
- The level stationary test shows only the stock price of Thailand is mean-reverting.
- The trend stationary test shows the stock prices are all mean-reverting.

This paper investigates whether stock prices in seven Asian stock markets can be characterized as random walk or mean reversion processes over the period December 1990 to March 2013. A Lagrange Multiplier (LM) Fourier unit root test proposed by Enders and Lee (2012) and a stationary test with a Fourier function proposed by Becker et al. (2006) are employed to approximate smooth structural breaks. The results of LM Fourier unit root test show that stock prices in all these seven stock markets demonstrate significant mean reversion. The results of the stationary test can be divided into two parts. When a level term is included in the model, the stock price of Thailand is mean-reverting while stock prices of other regions (Mainland China, HK, China, Japan, South Korea, Malaysia and Singapore) are non-stationary; when a trend term is included the stock prices in seven stock markets are all mean reversion processes.