On the validity of conventional statistical tests given evidence of non-synchronous trading and non-linear dynamics in returns generating process

Abstract

Based on the recent developments in market microstructure and applications of nonlinear dynamics and chaos theory to financial time series, the subsequent article questions the validity of traditional methods used to test the efficient market hypothesis. In particular, it emphasizes the invalidity of unit roots tests since they are not predictability tests.

Are Chinese stock markets efficient? Further evidence from a battery of nonlinearity tests

“…A recent article by Saadi et al (2006) has questioned the validity of those conventional statistical tests for testing EMH, and their criticisms applied to all earlier Chinese studies, since the efficiency of SHSE and SZSE was empirically examined using statistical tests that are designed to uncover linear correlations of price changes. Specifically, the statistical inferences of efficiency/inefficiency drawn from those conventional efficiency tests are on very shaky grounds.…”

“…In a recent article, Saadi et al (2006) questioned the validity of these conventional statistical tests for testing EMH on three grounds. First, a number of studies have shown that institutional features such as thin trading may generate spurious linear dependence in returns data.…”

“…Misleading conclusion could be reached when the underlying series have zero autocorrelation, yet possess predictable nonlinearities in mean, such as, those generated by bilinear and nonlinear moving average processes. Due to the above concern, Saadi et al (2006) advocated the use of statistical tests that are capable of detecting both linear and nonlinear structures in financial series such as the popular BDS (Brock, Dechert and Scheinkman) test of Brock et al (1996) in the empirical testing of market efficiency. Though overwhelming evidence in support of nonlinearity has been documented across international stock markets with different market structure mechanisms, indicating that the observed feature is a stylized fact of real financial data (see references cited in Lim et al, 2006), no study has formally tested the presence of nonlinearity in the Chinese stock markets except Seddighi and Nian (2004).…”

“…Having stated so, unless our allegation on the reliability of these studies has been verified empirically, one has to give them the benefit of the doubt at this stage. The above issues raised by Saadi et al (2006) are indeed pertinent, and hence, the efficiency of both the SHSE and SZSE warrants a re-examination. Heeding the advice of these authors to employ statistical tests that are capable of detecting both linear and nonlinear structures in financial series, the present article utilizes the 'nonlinearity toolkit' of Patterson and Ashley (2000) consisting of the McLeod-Li test (McLeod and Li, 1983), Engle Lagrange Multiplier (LM) test (Engle, 1982), BDS test (Brock et al, 1996), Tsay test (Tsay, 1986), Hinich bispectrum test (Hinich, 1982) and Hinich bicorrelation test (Hinich, 1996).…”

“…

Given that the efficiency of the Chinese stock markets was empirically examined in extant literature using statistical tests that are designed to uncover linear correlations of price changes, the obtained statistical inferences of efficiency/inefficiency are on very shaky grounds as highlighted in a recent article by Saadi et al (2006). Motivated by this concern, the present article re-examines the efficiency of the A-and B-shares markets in Shanghai and Shenzhen Stock Exchanges (SHSE and SZSE) using a battery of nonlinearity tests.

…”