Research on Decision-Making of Complex Venture Capital Based on Financial Big Data Platform

The prediction of stock premium has always been a hot issue. By predicting stock premiums to provide a way for companies to respond to financial risk investments, companies can avoid investment failures. In this paper, under the financial big data platform, bootstrap resampling technology and long short-term memory (LSTM) are used to predict the value of the stock premium within 20 months. First, using the theme crawler, jsoup page parsing, Solr search, and Hadoop architecture to build a platform for financial big data. Secondly, based on the block bootstrap resampling technology, the existing data information is expanded to make full use of the existing data information. Then, based on the LSTM network, the stock premium in 20 months is predicted and compared with the values predicted by support vector machine regression (SVR), and the SSE and R-square average indicators are calculated, respectively. The calculation results show that the SSE value of LSTM is lower than SVR, and the R-square value of LSTM is higher than SVR, which means that the effect of LSTM prediction is better than SVR. Finally, based on the forecast results and evaluation indicators of the stock premium, we provide countermeasures for the company’s financial risk investment.

[1]  Ai-Min Yang,et al.  Research on a Fusion Scheme of Cellular Network and Wireless Sensor for Cyber Physical Social Systems , 2018, IEEE Access.

[2]  Kiho Jeong,et al.  Recurrent support vector regression for a non-linear ARMA model with applications to forecasting financial returns , 2015, Comput. Stat..

[3]  Efstathios Paparoditis,et al.  Tapered block bootstrap , 2001 .

[4]  Vladimir Zadorozhny,et al.  Optimization Strategies for A/B Testing on HADOOP , 2013, Proc. VLDB Endow..

[5]  Ching-Hsue Cheng,et al.  A hybrid model based on rough sets theory and genetic algorithms for stock price forecasting , 2010, Inf. Sci..

[6]  Hiroshi Akima,et al.  A Method of Bivariate Interpolation and Smooth Surface Fitting for Irregularly Distributed Data Points , 1978, TOMS.

[7]  Jürgen Schmidhuber,et al.  Learning to Forget: Continual Prediction with LSTM , 2000, Neural Computation.

[8]  Roman M. Balabin,et al.  Support vector machine regression (SVR/LS-SVM)--an alternative to neural networks (ANN) for analytical chemistry? Comparison of nonlinear methods on near infrared (NIR) spectroscopy data. , 2011, The Analyst.

[9]  K. Singh,et al.  On the Asymptotic Accuracy of Efron's Bootstrap , 1981 .

[10]  Yi Zuo,et al.  Stock price forecast using Bayesian network , 2012, Expert Syst. Appl..

[11]  Chungang Yan,et al.  Big data resource service platform for the internet financial industry , 2014 .

[12]  Jürgen Schmidhuber,et al.  LSTM: A Search Space Odyssey , 2015, IEEE Transactions on Neural Networks and Learning Systems.

[13]  Pei-Chann Chang,et al.  Evolving and clustering fuzzy decision tree for financial time series data forecasting , 2009, Expert Syst. Appl..

[14]  Steven Walczak,et al.  An Empirical Analysis of Data Requirements for Financial Forecasting with Neural Networks , 2001, J. Manag. Inf. Syst..

[15]  Geoffrey R. John,et al.  Prediction of Tanzanian Energy Demand using Support Vector Machine for Regression (SVR) , 2015 .

[16]  Chen Wang,et al.  Improved v -Support vector regression model based on variable selection and brain storm optimization for stock price forecasting , 2016, Appl. Soft Comput..

[17]  Joshua D. Woodard,et al.  Big data and Ag-Analytics , 2016 .

[18]  Mario Vanhoucke,et al.  Support Vector Machine Regression for project control forecasting , 2014 .

[19]  S. Lahiri.  Theoretical comparisons of block bootstrap methods , 1999 .