兰州大学机构库 >数学与统计学院
基于人工智能算法的上海银行间同业拆放利率预测
Alternative TitleThe Artificial-intelligence-algorithms-based Prediction about the Shanghai Interbank Offered Rate
林庆添
Thesis Advisor严定琪
2016-05-14
Degree Grantor兰州大学
Place of Conferral兰州
Degree Name硕士
Keyword拆放利率 人工智能 SVR 布谷鸟搜索 粒子群优化
Abstract同业拆放利率是银行同业间在货币市场上融通资金的利率。作为市场化核心利率,上海银行间同业拆放利率(Shibor)能准确、及时地反映货币市场的资金供求关系,其变动会迅速传导和影响我国金融市场。因此,预测Shibor的波动和走势具有重要意义。 本文从两个时间维度上对Shibor的隔夜品种进行研究和预测,分别是日波动和月走势。隔夜Shibor日数据的预测中,首先建立基于BP神经网络算法的对照性预测模型,接着将小波神经网络应用于预测中,发现效果更好。进一步提出基于布谷鸟搜索(CS)优化的小波神经网络预测模型,提高了预测精度,较好地拟合了隔夜Shibor的日波动。预测隔夜Shibor的月均值时,从影响利率走势的因素出发,预选取9个指标并进行相关性检验,建立回归型支持向量机(SVR)预测模型。利用粒子群优化算法(PSO)改进SVR算法,建立PSO-SVR预测模型,提高预测精度,模型基本可以对隔夜Shibor的走势进行预测。 针对日数据和月数据的特点,选用相应适宜的算法,并进一步优化算法。建立的预测模型在隔夜Shibor的日数据和月均值预测中具有一定的科学性和应用价值。最后建立隔夜Shibor决策系统,整合预测模型,对货币市场参与主体具有一定指导作用。
Other AbstractInter Bank Offered Rate refers to the benchmark rate that banks use in circulating funds in the monetary market. As the core rate of the marketization, the Shanghai Interbank Offered Rate (Shibor) can accurately and timely reflect the supply and demand of funds in the monetary market and its variation will quickly influence the monetary market in China. Therefore, it is very important to predict the fluctuation and the tendency of the Shanghai Interbank Offered Rate. This paper will research and predict the overnight SHIOR varieties in two dimensions of time, which are the daily fluctuation and the monthly tendency. In order to predict the overnight Shibor daily data, firstly, establish a comparative prediction model based on Back-Propagation (BP) neural network algorithm. Secondly, apply the wavelet neural network to the prediction, which results in better effectiveness. Lastly, bring up the idea of wavelet neural network prediction model based on optimization of the Cuckoo Search (CS), which enhances the accuracy of the prediction and matches the daily fluctuation of the overnight Shibor in a better way. When predicting the monthly average of the overnight Shibor, from the point of the factors that will affect the tendency of the rate, select 9 indicators to inspect the correlation, and build the regressive Support Vector Regression (SVR) prediction model. After that, improve SVR algorithm by utilizing the Particle Swarm Optimization (PSO) and build the PSO-SVR prediction model to help enhance the accuracy of the prediction, in which way the model can basically predict the tendency of the overnight Shibor. Select and optimize the appropriate algorithm based on the features of the daily and monthly data. The built prediction models are scientific and applicable in some way in predicting the daily data and the monthly average rate of the overnight Shibor. Establishing the overnight Shibor deciding system and integrating the prediction models will certainly provide some guidance for the participants of the monetary market.
URL查看原文
Language中文
Document Type学位论文
Identifierhttps://ir.lzu.edu.cn/handle/262010/225136
Collection数学与统计学院
Recommended Citation
GB/T 7714
林庆添. 基于人工智能算法的上海银行间同业拆放利率预测[D]. 兰州. 兰州大学,2016.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Altmetrics Score
Google Scholar
Similar articles in Google Scholar
[林庆添]'s Articles
Baidu academic
Similar articles in Baidu academic
[林庆添]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[林庆添]'s Articles
Terms of Use
No data!
Social Bookmark/Share
No comment.
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.