HUANG Hongyun,WU Libin*,LI Shizheng,et al.Application of an improved SA-SVM algorithm injudgment of transaction signals[J].Journal of Yanbian University,2017,43(01):25-33.
一种改进的模拟退火算法优化的支持向量机在交易信号研判中的应用
- Title:
- Application of an improved SA-SVM algorithm in judgment of transaction signals
- 关键词:
- 模拟退火; 支持向量机; 择时; 三交换; LIBSVM ToolBox
- Keywords:
- simulated annealing; support vector machine; selection time; triple exchange method; LIBSVM ToolBox
- 分类号:
- TP183; F830.91; O29
- 文献标志码:
- A
- 摘要:
- 针对利用机器学习算法支持向量机来研判交易信号时的参数设置问题,本文首先提出一种改进的模拟退火算法(三交换法)来优化LIBSVM工具箱中的惩罚因子“-c”和核函数因子“-g”的选取,然后建立一个基于历史证券技术指标信息的量化择时模型.实证研究表明,改进后的ISA-SVM算法相比于二交换法和互逆交换法不仅可以更好地收敛于最小能量,而且在实际投资中可以对交易信号进行更准确的预测、实现更为可观的收益回报.
- Abstract:
- For solving the problem of parameter setting when using the machine learning algorithm Support Vector machine to judge the transaction signal, this paper firstly proposed an improved simulated annealing algorithm(triple exchange method)to optimize the selection of the penalty factor “-c” and the kernel function “-g” in the LIBSVM Toolbox, then establishing a quantitative timing model based on historical securities technical indicators. The empirical study shows that the improved ISA-SVM algorithm can not only converge to the minimum energy better than the two-exchange method and the inverse exchange method, but also can predict the transaction signal more accurately in the actual investment.
参考文献/References:
[1] 方浩文.量化投资发展趋势及其对中国的启示[J].管理现代化,2012(5):3-5.
[2] 李子睿.量化投资交易策略研究[D].天津:天津大学,2013.
[3] 章宏帆.运用量化投资策略实现超额收益Alpha的理论与实践[D].浙江大学,2015.
[4] 王力弘.浅议量化投资发展趋势及其对中国的启示[J].中国投资,2013(S2):202.
[5] 王冰,李想.浅议量化投资在国内市场的发展[J].经济视角(下),2011(3):46-47.
[6] 黄恒秋.基于高频数据的支持向量机量化择时预测模型[J].科技经济导刊,2016,(13):29.
[7] 苏冰.基于PCA-SVM模型的量化择时研究[D].天津财经大学,2015.
[8] 张伟.支持向量分类机(SVC)在量化选股中的应用[D].山东大学,2014.
[9] 李艺.量化交易策略模型应用研究[D].云南财经大学,2014.
[10] Corinna Cortes, Vladimir Vapnik. Support-vector networks[J]. Machine Learning, 1995,20(3):273-297.
[11] Olivier Chapelle, Vladimir Vapnik, Olivier Bousquet, et al. Choosing multiple parameters for support vector machines[J]. Machine Learning, 2002,46(1):131-159.
[12] 赵长春,姜晓爱,金英汉.非线性回归支持向量SMO算法改进[J].北京航空航天大学学报,2014(1):125-130.
[13] Cao L J, Keerthi S S, Ong Chong-Jin, et al. Parallel sequential minimal optimization for the training of support vector machines[J]. IEEE Transactions on Neural Networks, 2006,17(4):1039-1049.
[14] 薛浩然,张珂珩,李斌,等.基于布谷鸟算法和支持向量机的变压器故障诊断[J].电力系统保护与控制,2015,43(8):8-13.
[15] 庞峰.模拟退火算法的原理及算法在优化问题上的应用[D].吉林:吉林大学,2006.
[16] 朱颢东,钟勇.一种改进的模拟退火算法[J].计算机技术与发展,2009,19(6):32-35.
[17] 卢宇婷,林禹攸,彭乔姿,等.模拟退火算法改进综述及参数探究[J].大学数学,2015,31(6):96-103.
[18] 黄宏运,吴礼斌,李诗争.GA优化的SVM在量化择时中的应用[J].南京师范大学(工程技术版),2017,17(1):1-8.
相似文献/References:
[1]郭晓慧.基于LDA主题模型的文本语料情感分类改进方法[J].延边大学学报(自然科学版),2018,44(03):266.
GUO Xiaohui.The improved method based on LDA topic model foremotion classification of text corpus[J].Journal of Yanbian University,2018,44(01):266.
[2]黄永华,林振衡,陈学军.一种基于SVM的刹车蹄块片摩擦块表面裂纹检测法[J].延边大学学报(自然科学版),2019,45(02):175.
HUANG Yonghua,LIN Zhenheng,CHEN Xuejun.A surface crack detection method for frictionblock of brake pad based on SVM[J].Journal of Yanbian University,2019,45(01):175.
[3]俞颖,黄风华,阮奇.基于改进粒子群优化算法和CRNN的多类SVM分类[J].延边大学学报(自然科学版),2019,45(03):215.
YU Ying,HUANG Fenghua,RUAN Qi.Classification of multi-class support vector machines based on improved particle swarm optimization and CRNN[J].Journal of Yanbian University,2019,45(01):215.
[4]孙越,洪义成,刘鑫,等.SARIMA - SVR混合模型在电费收入预测中的应用[J].延边大学学报(自然科学版),2021,47(04):324.
SUN Yue,HONG Yicheng,LIU Xin,et al.Application of SARIMA - SVR hybrid model in electricity revenue forecasting[J].Journal of Yanbian University,2021,47(01):324.
备注/Memo
收稿日期: 2016-10-21 *通信作者: 吴礼斌(1964—),男,副教授,研究方向为计量金融与数理统计.
基金项目: 国家自然科学基金资助项目(11601001); 安徽高等学校省级自然科学基金资助项目(KJ2013Z001); 安徽财经大学校级重点研究项目(ACKY1402ZD)