We consider the viability of a modularised mechanistic online machine learning framework to learn signals in low-frequency financial time series data. The framework is proved on daily sampled closing time-series data from JSE equity markets. The input patterns are vectors of pre-processed sequences of daily, weekly and monthly or quarterly sampled feature changes. The data processing is split into a batch processed step where features are learnt using a stacked autoencoder via unsupervised learning, and then both batch and online supervised learning are carried out using these learnt features, with the output being a point prediction of measured time-series feature fluctuations. Weight initializations are implemented with restricted Boltzmann machine pre-training, and variance based initializations. Historical simulations are then run using an online feedforward neural network initialised with the weights from the batch training and validation step. The validity of results are considered under a rigorous assessment of backtest overfitting using both combinatorially symmetrical cross validation and probabilistic and deflated Sharpe ratios. Results are used to develop a view on the phenomenology of financial markets and the value of complex historical data-analysis for trading under the unstable adaptive dynamics that characterise financial markets.
In this note, we introduce how to use Volatility Index (VIX) for postprocessing quantitative strategies so as to increase the Sharpe ratio and reduce trading risks. The signal from this procedure is an indicator of trading or not on a daily basis. Finally, we analyze this procedure on SH510300 and SH510050 assets. The strategies are evaluated by measurements of Sharpe ratio, max drawdown, and Calmar ratio. However, there is always a risk of loss in trading. The results from the tests are just examples of how the method works; no claim is made on the suggestion of real market positions.
With the application of artificial intelligence in the financial field, quantitative trading is considered to be profitable. Based on this, this paper proposes an improved deep recurrent DRQN-ARBR model because the existing quantitative trading model ignores the impact of irrational investor behavior on the market, making the application effect poor in an environment where the stock market in China is non-efficiency. By changing the fully connected layer in the original model to the LSTM layer and using the emotion indicator ARBR to construct a trading strategy, this model solves the problems …
Sun Yat-sen University（中山大学）；Microsoft Research（微软研究院）
Successful quantitative investment usually relies on precise predictions of the future movement of the stock price. Recently, machine learning based solutions have shown their capacity to give more accurate stock prediction and become indispensable components in modern quantitative investment systems. However, the i.i.d. assumption behind existing methods is inconsistent with the existence of diverse trading patterns in the stock market, which inevitably limits their ability to achieve better stock prediction performance. In this paper, we propose a novel architecture, Temporal Routing Adaptor…
With the development of artificial intelligence technology, quantitative trading systems represented by reinforcement learning have emerged in the stock trading market. The authors combined the deep Q network in reinforcement learning with the sentiment quantitative indicator ARBR to build a high-frequency stock trading model for the share market. To improve the performance of the model, the PCA algorithm is used to reduce the dimensionality feature vector while incorporating the influence of market sentiment on the long-short power into the spatial state of the trading model and uses the LSTM…
To ensure reliable operation of power grids, their frequency shall stay within strict bounds. Multiple sources of disturbances cause fluctuations of the grid frequency, ranging from changing demand over volatile feed-in to energy trading. Here, we analyze frequency time series from the continental European grid in 2011 and 2017 as a case study to isolate the impact of trading. We find that trading at typical trading intervals such as full hours modifies the frequency fluctuation statistics. While particularly large frequency deviations in 2017 are not as frequent as in 2011, large deviations a…
This paper presents a quantitative analysis of the relationship between the stock market returns and corresponding trading volumes using high- frequency data from the Polish stock market. First, for stocks that were traded for suffciently long period of time, we study the return and volume distributions and identify their consistency with the power-law functions. We find that, for majority of stocks, the scaling exponents of both distri- butions are systematically related by about a factor of 2 with the ones for the returns being larger. Second, we study the empirical price impact of trades of…
Amundi Asset Management（东方汇理资产管理公司）；Ecole des Ponts ParisTech（巴黎高科路桥学校）
This article explores the use of machine learning models to build a market generator. The underlying idea is to simulate artificial multi-dimensional financial time series, whose statistical properties are the same as those observed in the financial markets. In particular, these synthetic data must preserve the probability distribution of asset returns, the stochastic dependence between the different assets and the autocorrelation across time. The article proposes then a new approach for estimating the probability distribution of backtest statistics. The final objective is to develop a framewo…
We study the intraday behaviour of the statistical moments of the trading volume of the blue chip equities that composed the Dow Jones Industrial Average index between 2003 and 2014. By splitting that time interval into semesters, we provide a quantitative account of the non-stationary nature of the intraday statistical properties as well. Explicitly, we prove the well-known U-shape exhibited by the average trading volume-as well as the volatility of the price fluctuations-experienced a significant change from 2008 (the year of the sub-prime financial crisis) onwards. That has resulted in a fa…
Using high-frequency time series of stock prices and share volumes sizes from January 2002-May 2009, this paper investigates whether the effects of the onset of high-frequency trading, most prominent since 2005, are apparent in the dynamics of the dollar traded volume. Indeed it is found in almost all of 14 heavily traded stocks, that there has been an increase in the Hurst exponent of dollar traded volume from Gaussian noise in the earlier years to more self-similar dynamics in later years. This shift is linked both temporally to the Reg NMS reforms allowing high-frequency trading to flourish…
Studying the micro-trading behaviors before stock price jumps is an important problem for financial regulations and investment decisions. In this study, we provide a new framework to study pre-jump trading behaviors based on multivariate time series analysis. Different from the existing literature, our methodology takes into account the temporal information embedded in the trading-related attributes and can better evaluate and compare the abnormality levels of different attributes. Moreover, it can explore the joint informativeness of the attributes as well as select a subset of highly informa…
Technical trading represents a class of investment strategies for Financial Markets based on the analysis of trends and recurrent patterns of price time series. According standard economical theories these strategies should not be used because they cannot be profitable. On the contrary it is well-known that technical traders exist and operate on different time scales. In this paper we investigate if technical trading produces detectable signals in price time series and if some kind of memory effect is introduced in the price dynamics. In particular we focus on a specific figure called supports…
Employing a recent technique which allows the representation of nonstationary data by means of a juxtaposition of locally stationary patches of different length, we introduce a comprehensive analysis of the key observables in a financial market: the trading volume and the price fluctuations. From the segmentation procedure we are able to introduce a quantitative description of a group of statistical features (stylizes facts) of the trading volume and price fluctuations, namely the tails of each distribution, the U-shaped profile of the volume in a trading session and the evolution of the tradi…
Based on 1-minute price changes recorded since year 2012, the fluctuation properties of the rapidly-emerging Bitcoin (BTC) market are assessed over chosen sub-periods, in terms of return distributions, volatility autocorrelation, Hurst exponents and multiscaling effects. The findings are compared to the stylized facts of mature world markets. While early trading was affected by system-specific irregularities, it is found that over the months preceding Apr 2018 all these statistical indicators approach the features hallmarking maturity. This can be taken as an indication that the Bitcoin market…
In this report, we talked about a new quantitative strategy for choosing the optimal(s) stock(s) to trade. The basic notions are generally very known by the financial community. The key here is to understand 1) the standard score applied to a sample and 2) the correlation factor applied to different time series in real life. These notions are the core of our research. We are going to begin with the introduction section. In this part, we talked about variance, covariance, correlation factor, daily returns in stock trading and the Shapiro-Wilk test to test the normality of a time serie. Next to …
We study the cascading dynamics immediately before and immediately after 219 market shocks. We define the time of a market shock T_{c} to be the time for which the market volatility V(T_{c}) has a peak that exceeds a predetermined threshold. The cascade of high volatility "aftershocks" triggered by the "main shock" is quantitatively similar to earthquakes and solar flares, which have been described by three empirical laws --- the Omori law, the productivity law, and the Bath law. We analyze the most traded 531 stocks in U.S. markets during the two-year period 2001-2002 at the 1-minute time res…
Unveiling the Potential of Sentiment: Can Large Language Models Predict Chinese Stock Price Movements?
International Digital Economy Academy (IDEA)（国际数字经济学院（IDEA））；The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Data Science and Analytics Thrust, The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州）数据科学与分析学域）；Financial Technology Thrust, The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州）金融科技学域）
Quantum theory is used to model secondary financial markets. Contrary to stochastic descriptions, the formalism emphasizes the importance of trading in determining the value of a security. All possible realizations of investors holding securities and cash is taken as the basis of the Hilbert space of market states. The temporal evolution of an isolated market is unitary in this space. Linear operators representing basic financial transactions such as cash transfer and the buying or selling of securities are constructed and simple model Hamiltonians that generate the temporal evolution due to c…
Unlike price fluctuations, the temporal structure of cryptocurrency trading has seldom been a subject of systematic study. In order to fill this gap, we analyse detrended correlations of the price returns, the average number of trades in time unit, and the traded volume based on high-frequency data representing two major cryptocurrencies: bitcoin and ether. We apply the multifractal detrended cross-correlation analysis, which is considered the most reliable method for identifying nonlinear correlations in time series. We find that all the quantities considered in our study show an unambiguous …
We consider different levels of complexity which are observed in the empirical investigation of financial time series. We discuss recent empirical and theoretical work showing that statistical properties of financial time series are rather complex under several ways. Specifically, they are complex with respect to their (i) temporal and (ii) ensemble properties. Moreover, the ensemble return properties show a behavior which is specific to the nature of the trading day reflecting if it is a normal or an extreme trading day.
