Predicting the trend of stock index based on feature engineering and CatBoost model

As an important tool to measure the current situation of the whole stock market, the stock index has always been the focus of researchers, especially for its prediction. This paper uses trend types, which are received by clustering price series under multiple time scale, combined with the day-of-the-week effect to construct a categorical feature combination. Based on the historical data of six kinds of Chinese stock indexes, the CatBoost model is used for training and predicting. Experimental results show that the out-of-sample prediction accuracy is 0.55, and the long–short trading strategy can obtain average annualized return of 34.43%, which is a great improvement compared with other classical classification algorithms. Under the rolling back-testing, the model can always obtain stable returns in each period of time from 2012 to 2020. Among them, the SSESC’s long–short strategy has the best performance with an annualized return of 40.85% and a sharp ratio of 1.53. Therefore, the trend information on multiple time-scale features based on feature engineering can be learned by the CatBoost model well, which has a guiding effect on predicting stock index trends.

Portfolio Construction of Energy-Related Assets

Portfolio construction is one of the most fatal issues of modern finance, which can effectively gain returns or reduce risks. This study constructs portfolios in energy-related assets. Specifically, the Monte Carlo simulations are carried out for a hundred thousand times in order to discover the efficient frontier and find the minimum variance and the maximum sharp ratio portfolio. According to the simulations, the American Electric Power possesses the largest share in minimum variance portfolio, while NextEra Energy for sharp ratio method. The results may benefit certain investor in financial markets and shed lights to focus more on portfolio allocation during constructing.

Using Data Augmentation Based Reinforcement Learning for Daily Stock Trading

In algorithmic trading, adequate training data set is key to making profits. However, stock trading data in units of a day can not meet the great demand for reinforcement learning. To address this problem, we proposed a framework named data augmentation based reinforcement learning (DARL) which uses minute-candle data (open, high, low, close) to train the agent. The agent is then used to guide daily stock trading. In this way, we can increase the instances of data available for training in hundreds of folds, which can substantially improve the reinforcement learning effect. But not all stocks are suitable for this kind of trading. Therefore, we propose an access mechanism based on skewness and kurtosis to select stocks that can be traded properly using this algorithm. In our experiment, we find proximal policy optimization (PPO) is the most stable algorithm to achieve high risk-adjusted returns. Deep Q-learning (DQN) and soft actor critic (SAC) can beat the market in Sharp Ratio.

Vector Autoregressive Weighting Reversion Strategy for Online Portfolio Selection

Aiming to improve the performance of existing reversion based online portfolio selection strategies, we propose a novel multi-period strategy named “Vector Autoregressive Weighting Reversion” (VAWR). Firstly, vector autoregressive moving-average algorithm used in time series prediction is transformed into exploring the dynamic relationships between different assets for more accurate price prediction. Secondly, we design the modified online passive aggressive technique and advance a scheme to weigh investment risk and cumulative experience to update the closed-form of portfolio. Theoretical analysis and experimental results confirm the effectiveness and robustness of our strategy. Compared with the state-of-the-art strategies, VAWR greatly increases cumulative wealth, and it obtains the highest annualized percentage yield and sharp ratio on various public datasets. These improvements and easy implementation support the practical applications of VAWR.

An Investigation of the Return Risk and Liquidity Measure for Chinese Open-Ended Funds

So far, the fund industry has become one of the four backbones of the Chinese financial system, together with the banking industry, the security industry and the insurance industry. In addition, open-ended fund shares are the mainstream of the fund industry, and the product characteristics and operational characteristics of open-ended funds will lead to an unavoidable risk of return with liquidity measure. Therefore, based on the theories of financial investment, this research profile the risk return and liquidity characteristics of three different open-ended funds in China, which are helpful to making rational investments. This article selects three different kinds of funds with the relevant data from 2012to 2017 from the Huaxia Fund Management Co. Ltd., for each fund, the authors report the beta, Sharp Ratio, Information ratio and illiquidity. These risk-return features are discussed in the context of the different asset classes that each fund has invested, thus eventually obtaining a fund which has smaller relative liquidity risk and higher return after comparing. Hence, the investor can make the rational investment from the analysis of empirical results.

An Investigation of the Return Risk and Liquidity Measure for Chinese Open-Ended Funds

So far, the fund industry has become one of the four backbones of the Chinese financial system, together with the banking industry, the security industry and the insurance industry. In addition, open-ended fund shares are the mainstream of the fund industry, and the product characteristics and operational characteristics of open-ended funds will lead to an unavoidable risk of return with liquidity measure. Therefore, based on the theories of financial investment, this research profile the risk return and liquidity characteristics of three different open-ended funds in China, which are helpful to making rational investments. This article selects three different kinds of funds with the relevant data from 2012to 2017 from the Huaxia Fund Management Co. Ltd., for each fund, the authors report the beta, Sharp Ratio, Information ratio and illiquidity. These risk-return features are discussed in the context of the different asset classes that each fund has invested, thus eventually obtaining a fund which has smaller relative liquidity risk and higher return after comparing. Hence, the investor can make the rational investment from the analysis of empirical results.

Two Stage Portfolio Selection and Optimization Model with the Hybrid Particle Swarm Optimization

The selection criteria play an important role in the portfolio optimization using any ratio model. In this paper, the authors have considered the mean return as profit and variance of return as risk on the asset return as selection criteria, as the first stage to optimize the selected portfolio. Furthermore, the sharp ratio (SR) has been considered to be the optimization ratio model. In this regard, the historical data taken from Shanghai Stock Exchange (SSE) has been considered. A metaheuristic technique has been developed, with financial tool box available in MATLAB and the particle swarm optimization (PSO) algorithm. Hence, called as the hybrid particle swarm optimization (HPSO) or can also be called as financial tool box particle swarm optimization (FTB-PSO). In this model, the budgets as constraint, where as two different models i.e. with and without short sale, have been considered. The obtained results have been compared with the existing literature and the proposed technique is found to be optimum and better in terms of profit.

QUANTIFYING THE IMPACT OF PARTIAL INFORMATION ON SHARPE RATIO OPTIMIZATION

Motivated by the fact that many investors have limited ability to update the expectation regarding future stock returns with the arrival of new information instantly, this paper provides a continuous-time model to study the performance of passive trading strategies. We derive the true Sharp ratio of the passive strategies in terms of the mean and variance of an explicit stochastic process. Based on this expression, we quantify the impact of partial information by performing a thorough comparative static analysis. Such an analysis provides a rationale for why investors with inaccurate information about stock return behave better in the mean-reverting environment than in the i.i.d. environment and why pessimistic investors can achieve better performance than optimistic ones. As a by-product, we propose an analytical approach to compute the “implied” parameters in stock return predictor for both i.i.d. and mean-reverting dynamics, which seems interesting for future research.

Modeling and Trading the EUR/USD Exchange Rate Using Machine Learning Techniques

The present paper aims in investigating the performance of state-of-the-art machine learning techniques in trading with the EUR/USD exchange rate at the ECB fixing. For this purpose, five supervised learning classification techniques (K-Nearest Neighbors algorithm, Naïve Bayesian Classifier, Artificial Neural Networks, Support Vector Machines and Random Forests) were applied in the problem of the one day ahead movement prediction of the EUR/USD exchange rate with only autoregressive terms as inputs. For comparison reasons, the performance of all machine learning techniques was benchmarked by two traditional techniques (Naïve Strategy and moving average convergence/divergence model). Trading strategies produced by the machine learning techniques of Support Vector Machines and Random Forests clearly outperformed all other strategies in terms of annualized return and sharp ratio. To the best of our knowledge, this is the first application of Random Forests in the problem of trading with the EUR/USD exchange rate providing extremely satisfactory results.