Stability of Markovian structure observed in high frequency foreign exchange data

Abstract

Abstract. Contrary to the common sense in economics and financial engineering, price fluctuations at very fine level of motion exhibit various evidences against the efficient market hypothesis. We attempt to investigate this issue by studying extensive amount of foreign currency exchange data for over five years at the finest level of resolution. We specifically focus on the proposed stability in binomial conditional probabilities originally found in much smaller examples of financial time series. In order to handle very large data, we have written an efficient program in C that automatically generates those conditional probabilities. It is found that the stability is maintained for extremely large time duration that covers almost the entire period. Based on the length of conditions for which the conditional probabilities are distinguishable each other, we identify the length of memory being less than 3 movements.

Machine-learning classification techniques for the analysis and prediction of high-frequency stock direction

“…Ohira et al [171], Tanaka-Yamawaki [212], Sazuka [199], and Hashimoto et al [100] examined market data at the lowest intra-day level available, trade-by-trade (sometimes known as tick data) and found extremely high levels of predictability. For example, in [171] and [212] the authors report predictability as high as 79.7% and 75.0% respectively. While the movements are clearly predictable and raise doubt as to the efficiency of the currency market, we theorize here that much of the predictability in those two papers can be explained by the noisy continuation 4 of the bid-ask market dynamics 5 .…”

“…Since the stock's previous day's closing price and the current day's opening price are often different, conditional probabilities for individual days are computed separately. This discontinuity is a distinct disadvantage of equity data over currency data, such as was used in [171,212]. The worldwide nature of currency trading allows for the market to be continually open somewhere on Earth, except for weekends.…”

“…believe that the high levels of predictability found in [171,212] have more to do with the bid-ask spread than the past price change. We empirically examine the conditional probabilities of trade-by-trade (tick) data along with nine temporal timespans of 1, 3, 5, 10, and 30 seconds and 1, 5, and 30 minutes for 52 separate one-week periods in 2005 of a popularly traded stock, the Standard and Poor's 500 index (symbol: SPY).…”

Two Stochastic Phases of Tick-wise Price Fluctuation and the Price Prediction Generator

“…This fact motivates us to predict the immediate future of the financial time series based on its past movements. In order to bring such knowledge in to a practical use, we have applied earlier the method of evolutional strategy [4] to generate the prediction on the next move [5], by adaptively learning the pattern of price movement. We have found, however, a certain stochastic regularity exists in the tick-wise motion of price changes, namely, the direction of the price at the next tick goes to the opposite direction of the previous move in about 70%, which implies the possibility of short-term prediction [3].…”