Investigation of efficient market hypothesis with nonlinear Fourier Quantile unit root test: An application in OECD countries

The aim of this study is to test the existence of efficient markets hypothesis (EMH) in the stock markets of OECD (Organisation for Economic Cooperation and Development) countries. For each sample, the longest period is used as the analysis period, considering data availability. For this purpose, firstly we used, tests that can be categorised as conventional unit root tests in which the Augmented Dickey and Fuller (1981) (ADF) unit root test, Zivot and Andrews (1992) (ZA) unit root test, Im and Schmidt (2008) residuals-augmented least squares (RALS-ADF) unit root test, Kapetanios, Shin and Snell (2003) (KSS) unit root test, Enders and Lee (2012) Fourier (F-ADF) unit root test and Christopoulos and Leon-Ledesma (2010) non-linear Fourier (F-KSS) unit root test. Then, the Fourier Non-linear Quantile Unit Root Test (FNQKS) developed by Bahmani-Oskooee et al. (2020), which distinguishes this study from other studies in the literature, is used. According to the results of ADF unit root tests, in 5 countries, in the RALS-ADF test, in 6 countries, in the KSS test, in 13 countries, in the ZA test, in 14 countries, in the F-ADF test, in 11 countries, and in the F-KSS test, in 11 countries the null hypothesis of unit root is rejected , that is, EMH is not valid for the stock markets of these countries. According to the results of the FNQKS test, which distinguishes the study from other studies in the literature, the null hypothesis of unit root is rejected for 27 countries. Accordingly, contrary to the results obtained from conventional tests, FNQKS test results show that the EMH is invalid in approximately 70 per cent of OECD countries. Quantile-based results show that positive and negative shocks have different effects on the EPH in OECD countries, that is, there is an asymmetric structure.