Portfolio Selection with Support Vector Regression - R/Finance 2018

Portfolio Selection with Support Vector Regression. Henrique, Pedro Alexandre. University of Brasilia, Brazil. Machine Learning. SVM & SVR. Stocks selection.
展开查看详情

1.Portfolio Selection with Support Vector Regression Henrique, Pedro Alexandre University of Brasilia, Brazil

2. WHY SVM? • Machine Learning • SVM & SVR • Stocks selection Chih-Chung Chang and Chih-Jen Lin, LIBSVM : a library for support vector machines, 2001. • Multiple dimensions Expand the information from the variables From Dr.Sead Sayad web site -Support Vector The importance of choosing Machine - Regression (SVR) Kernel

3. APPLICATION • SVR – Support Vector • Test different 15 Regression Kernels for portfolio selection to beat the market The dual function: Kernel ( Multi dimensional mapping) Gaussian Radial Basis Kernel: Predict function.

4. WORKFLOW • S&P 100 – from 06/30/2014 • Fundamentalist analysis • Feature Selection • Fundamental data from 06/29/1990 • From 127 down to 24 to 06/30/2014. features Cross Validation Training Validation Test 52,5% 22,5% 25% Random Selection

5.STRATEGY • Forecasting the quarterly return of the stocks for the Portfolio Selections. • 15 portfolios - weighted by the forecast return • Benchmark for the portfolios: • Equal weighted portfolios with the 100 stocks.

6.RESULTS

7. 374,40% 192,65%

8.

9. RESEARCHERS IN PROGRESS • Machine per sector • Other inputs • Kernel combination • SVM with risk management tools

10. THANK YOU! Fan, A., & Palaniswami, M. (2001). Stock selection using support vector machines. Paper presented at the Neural Networks, 2001. Proceedings. IJCNN'01. International Joint Conference on. Marcelino, S., Henrique, P. A., & Albuquerque, P. H. M. (2015). Portfolio selection with support vector machines in low economic perspectives in emerging markets. Economic Computation & Economic Cybernetics Studies & Research, 49(4). Huerta, Ramon, Fernando Corbacho, and Charles Elkan. "Nonlinear support vector machines can systematically identify stocks with high and low future returns." Algorithmic Finance 2.1 (2013): 45-58. Emir, S., Dinçer, H., & Timor, M. (2012). A Stock Selection Model Based on Fundamental and Technical  Analysis Variables by Using Artificial Neural Networks and Support Vector Machines. Review of  Economics & Finance, 106­122. Packages: Robustbase PerformanceAnalytics Pedro Alexandre M.B. Henrique. Ggplot2 robustbase pedroalexandre.df@gmail.com Dplyr Scales Kernlab Fselector Mlbench Foreach doParallel doSNOW rgl