Academic Journal of Business & Management, 2023, 5(5); doi: 10.25236/AJBM.2023.050523.
Jinhui Zhang, Lanlan Shi, Wei Xu
School of Statistical and Applied Mathematics, Anhui University of Finance and Economics, Bengbu, 233030, Anhui, China
In today's trading market, for every investor, they want to obtain a higher rate of return within their own controllable risks. And how to obtain a maximum yield is undoubtedly a complex and difficult challenge. In order to help investors obtain a higher return in gold and bitcoin investment, we have integrated XGBoost, ESN, Black-Litterman and other models to establish a more optimized quantitative investment plan. Task 1: For the model, first use the optimized XGBoost-ESN model to predict the prices of gold and Bitcoin on the second day, and then use the Black-Litterman model to help investors make more accurate quantitative investments. From the initial $1,000 to the final $20,278.14, the annualized rate of return: 82.56%. Task 2: In order to predict the accuracy of the price, we use two different price prediction methods for prediction. The weights of the two are determined by the least squares method, and the optimized model is obtained. And in order to understand that our investment cycle is the optimal solution, we used different trading cycles of 5 days, 15 days, and 30 days for gold and Bitcoin to conduct a single-variable comparison test, and obtained different optimal transactions between the two cycle. Task 3: We conduct a dynamic analysis of transaction costs, and determine the optimal transaction cost range for gold and Bitcoin by understanding different transaction costs. Task 4: Since different investors have different risk acceptance levels, we have added var risk prediction content. In order to reduce the loss of income, we set a profit and loss line to ensure the safety of funds. Finally, we discuss the problems that may arise in the process of practical application of the model, and use the validity test to verify the validity of the established model. In addition, we provide a reasonable evaluation of the strengths and weaknesses of the model.
XGBoost, ESN, Black-Litterman, Quantitative Trading, Asset Allocation, VAR
Jinhui Zhang, Lanlan Shi, Wei Xu. Research on Optimization Strategy of Quantitative Investment Scheme Based on Black-Litterman Model. Academic Journal of Business & Management (2023) Vol. 5, Issue 5: 164-176. https://doi.org/10.25236/AJBM.2023.050523.
 Wang M , Guo C , Zhou Y , et al. Research on optimization operation strategy of building-type distributed energy system based on Ebsilon[C]// 2020 IEEE Sustainable Power and Energy Conference (iSPEC). IEEE, 2020
 Liu Keshi, Weng Tongfeng, Gu Changgui, Yang Huijie. (2020) Visibility Graph Analysis of Bitcoin Price Series, Physics A: Statistical Mechanics and Its Applications, 38(C)
 Chen Tingxuan, Chen Muyan, (2020) Du Guanting. Determinants of Bitcoin Price: Utilization of GARCH and Machine Learning Methods, Computational Economics, (pre-published)
 Naresh Gopal, K.S. Sentir Kumar. (2020) Predicting Bitcoin price - ANN method, International Journal of Electronic Finance, 10(1-2)
 Wang Yufang, Wang Haiyan. (2020) Use networks and partial differential equations to predict Bitcoin price movements, Chaos (Woodbury, NY), 30(7)
 Viebahn P, Vallentin D, Hoeller S. Prospects of carbon capture and storage (CCS) in China's power sector -n An integrated assessment [J]. Applied Energy, 2015, 157(NOV.1):229-244
 Schwartz J A. Managing the public sector research & development portfolio selection process: a case study of quantitative selection and optimization. 2016
 Divya Aggarwal, (2020) Shabana Chandrasekaran, Balamurugan Annamalai. A Bitcoin Price Prediction Method Based on Empirical Ensemble Pattern Decomposition and Support Vector Machines, Journal of Behavioral and Experimental Finance, 27 (pre-published)
 Shenkevich Andrei. (2020) Technical Analysis and Earning Predictability of Bitcoin Futures, Bitcoin Price, Forecast Magazine, 40(4)
 Li G, Li H, Zeng X, et al. Research on Optimization Control Strategy for Limiting Short-Circuit Current Based on Impedance Sensitivity and Fuzzy Quantitative Evaluation[J]. Journal of Shandong Electric Power College, 2012:2094-2098
 Jon Crowcroft, Damiano di Francisco Mesa, Alessandro Magrini, Andrea Marino, Laura Ricci. (2020) Analysis of Bitcoin Price Using User Graph and Trusted Transactions, IEEE Transactions on Network Science and Engineering
 Bulcu Capal, Jose Olmo. (2020) Analysis of Bitcoin Price Using Market and Sentiment VARiables, World Economy, 44(1)
 M. Poongodi, V. Vijayakumar, Naveen Chilamkurti. (2020) Bitcoin Price Prediction Based on ARIMA Model, International Journal of Internet Technology and Secured Transactions, 10(4)
 Yang Li, Zheng Zibin, Dai Hongning. (2020) Using Attentive LSTM and Embedding Network to Enhance Bitcoin Price Volatility Prediction, Applied Science, 10(14)
 Mahboubeh Faghih Mohammadi Jalali, Hanif Heidari. (2020) Using Grey System Theory to Predict Bitcoin Price Changes, Financial Innovation, 6(3)
 Chen Zheshi, Li Chunhong, Sun Wenjun. (2020) Bitcoin Price Prediction Based on Machine Learning: A Sample Dimension Engineering Method, Journal of Computational and Applied Mathematics, 365(C)
 Thomas Dimpfl, Stefania Odelli. (2020) Bitcoin Price Risk——Perspective of Duration, Journal of Risk and Financial Management, 13(7)