Search

• Study of the CTBLPO problem through Error-Correction Neural Networks. • Investor's views are regarded as a forecasting problem. • Proposal of a feed-forward MAWTSNN model for time-series forecasting problems. • Four other high-performing NN models are compared to the MAWTSNN model. • The use of TVQP NN solvers, i.e. ZNN and LVI-PDNN, to finance. • Experiments in three distinct portfolio configurations using real-world datasets. The Black-Litterman (BL) model is a particularly essential analytical tool for effective portfolio management in financial services sector since it enables investment analysts to integrate investor views into market equilibrium returns. In this research, we define and study the continuous-time BL portfolio optimization (CTBLPO) problem as a time-varying quadratic programming (TVQP) problem. The investor's views in the CTBLPO problem are regarded as a forecasting problem, and they are generated by a novel neural network (NN) model. More precisely, employing a novel multi-function activated by a weights-and-structure-determination for time-series (MAWTS) algorithm, a 3-layer feed-forward NN model, called MAWTSNN, is proposed for handling time-series modeling and forecasting problems. Then, using real-world datasets, the CTBLPO problem is approached by two different TVQP NN solvers. These solvers are the zeroing NN (ZNN) and the linear-variational-inequality primal–dual NN (LVI-PDNN). The experiment findings illustrate and compare the performances of the ZNN and LVI-PDNN in three various portfolio configurations, as well as indicating that the MAWTSNN is an excellent alternative to the traditional approaches. To promote and contend the outcomes of this research, we created two MATLAB repositories for the interested user, that are publicly accessible on GitHub. [ABSTRACT FROM AUTHOR]

• Fear/greed indicator predictions act a valuable proxy for future investor sentiment and constructing relative Black-Litterman views. • A hybrid CEEMDAN-GRU deep learning model gives the most accurate predictions for this indicator. • The XGBoost ensemble algorithm is used to predict relative outperformance in returns when constructing Black-Litterman views. • The proposed Black-Litterman approach outperforms eight other established portfolio strategies. We introduce a new dimension in constructing relative investor views for the Black-Litterman model by incorporating fear/greed technical indicator predictions as a proxy for investor sentiment in the portfolio construction process. We apply a hybrid CEEMDAN-GRU deep learning model to forecast this indicator and the XGBoost ensemble learning algorithm to forecast returns for ten country ETFs and create relative views for the Black-Litterman model. These models beat several benchmark forecasting models. Our empirical results show that the proposed approach outperforms the Markowitz, minimum-variance, equally-weighted and risk-parity strategies along with four other Black-Litterman approaches from the literature for six investment periods. [ABSTRACT FROM AUTHOR]

• CNN is efficient in feature extraction and BiLSTM is efficient in long-term multivariate time series prediction. • CNN-BiLSTM gives the best predictions when compared to CNN, LSTM, BiLSTM and CNN-LSTM. • The predictions from the CNN-BiLSTM model are used as investor views inside a Black-Litterman model with time variation in the conditional distribution of returns. • The Dynamic Black-Litterman model with CNN-BiLSTM predictions as views outperforms three benchmark models by a significant margin in terms of financial efficiency and diversification. We use daily price and technical indicators' data for the ten MSCI Asia Pacific sector indices for the past 20 years and find that our hybrid multivariate Convolutional Neural Network - Bidirectional Long Short-Term Memory (CNN-BiLSTM) deep learning model gives reasonably better predictions when predicting index closing prices out-of-sample than using either CNN or BiLSTM alone. After utilizing these predictions as investor views inside the Black-Litterman model with time variation in the conditional distribution of returns, we find that the portfolios generated outperform all benchmark model portfolios by a considerable margin in terms of financial efficiency and diversification. [ABSTRACT FROM AUTHOR]

• This paper introduces a deep Frequency decomposition model as a stock prediction model. • The role of introduced prediction model in the portfolio management is highlighted. • The Black-Litterman model is addressed as a proper response to the investors' subjective views. • The collaboration of the models above could make the intelligent asset allocation more effective. The boom of financial technology and artificial intelligence (AI) has influenced financial industry. Application of AI-based techniques especially deep neural networks in the stock market has drawn particular attention in recent years. Moreover, applying modern technology in finance has created new fields such as computational finance and intelligent asset management. Following this revolution, the present research aims to shed light on deep learning models in stocks analysis and apply theses analyses in portfolio management. To this end, we introduced a novel hybrid deep learning model for stock prediction and incorporated these predictions as investors' views in Black-Litterman asset allocation model. The finding indicates that the Black-Litterman portfolio based on the predictions of the hybrid CEEMD-CNN-LSTM model constructed portfolios with high return, low extreme allocation, and low risk. Furthermore, the Black-Litterman portfolios based on the introduced prediction model outperformed the mean-variance portfolio, equal-weighted portfolio, and the Black-Litterman portfolios based on the predictions of CNN-LSTM and LSTM. This remarkable performance could be attributed to each constituting model used in the hybrid model and portfolio formation strategy. [ABSTRACT FROM AUTHOR]