Noise traders: a new approach to understand the phantom of stock markets

Abstract

In microstructure literature, the ambiguity regarding the distinctive features of noise traders suggests that a further exploration of their behaviour is needed. In this study, we attempt to illuminate the intraday behavioural aspects of noise traders in the Turkish stock market using a novel approach. The diagnostic results in the initial phase of the analysis reveal that residual volume is a suitable proxy to identify noise traders' activities. The second phase of the analysis manifests that the noise traders significantly contribute to the volatility in spreads and that the duration of the volatility impact is short lived. These results are consistent with our prior conjectures and affirm the validity of our approach and propositions. The approach and findings, when generalized to other emerging markets, carry some implications for policy markers. In Turkey, revival of capital gain taxes abolished after the inception of global crisis might be a resolution to alleviate noise trader activities.

Notes

¹In intraday spread series, as in most of the intraday asset return patterns, a serial dependence is observed. In the mean equation, an appropriate number of autoregressive and moving average terms are added to the model to correct for the serial correlation problem in spread series.

²No-noise model represents the restricted version of Equation 4(4) where residual volume is not included. Thus, it represents the standard GARCH model.