Last modified: 2019-12-19
Abstract
We examine intraday security prices using Bloomberg market prices from October to December 2018. Our sample consists of 12,162,150 data points, covering 495 stocks, 63 trading days and 390 minutes per day. It is well known that news arrival impacts market prices. Casual empiricism suggests that news arrival is random and therefore intraday market prices may also be random. However, following Lo and Mackinlay (1988) and Chow and Denning (1993), we reject the linear process of randomness in intraday security prices. Following Webel (2011), we then examine the non-linear chaotic process using the zero-one test adjusted by wavelet denoising prior to testing. Our results are consistent with chaotic intraday security returns and inferentially with chaotic news arrival.