期刊文献+
共找到2篇文章
< 1 >
每页显示 20 50 100
基于评分统计预测的协同过滤算法
1
作者 周延鹏 张兴明 《信息工程大学学报》 2016年第6期646-650,693,共6页
针对没有考虑用户和项目的相似性,推荐精度较低的问题,提出评分统计预测模型,定义用户和项目统计信息分别表示用户个人喜好和项目品质;根据统计信息,产生符合正态分布随机数来表示项目品质评分和个人喜好的评分;最后结合两个评分建立线... 针对没有考虑用户和项目的相似性,推荐精度较低的问题,提出评分统计预测模型,定义用户和项目统计信息分别表示用户个人喜好和项目品质;根据统计信息,产生符合正态分布随机数来表示项目品质评分和个人喜好的评分;最后结合两个评分建立线性回归预测模型进行评分预测,并据此设计了相关算法。实验结果表明,文章提出的算法推荐精度高于传统协同过滤算法。 展开更多
关键词 协同过滤 评分统计预测 用户统计信息 项目统计信息
下载PDF
Calendar Effects in AAPL Value-at-Risk
2
作者 Hong-Kun Zhang Zijing Zhang 《Journal of Mathematics and System Science》 2016年第6期215-233,共19页
This study investigates calendar anomalies: day-of-the-week effect and seasonal effect in the Value-at-Risk (VaR) analysis of stock returns for AAPL during the period of 1995 through 2015. The statistical propertie... This study investigates calendar anomalies: day-of-the-week effect and seasonal effect in the Value-at-Risk (VaR) analysis of stock returns for AAPL during the period of 1995 through 2015. The statistical properties are examined and a comprehensive set of diagnostic checks are made on the two decades of AAPL daily stock returns. Combing the Extreme Value Approach together with a statistical analysis, it is learnt that the lowest VaR occurs on Fridays and Mondays typically. Moreover, high Q4 and Q3 VaR are observed during the test period. These results are valuable for anyone who needs evaluation and forecasts of the risk situation in AAPL. Moreover, this methodology, which is applicable to any other stocks or portfolios, is more realistic and comprehensive than the standard normal distribution based VaR model that is commonly used. 展开更多
关键词 Risk Measures VALUE-AT-RISK Extreme value theory Generalized Pareto Distribution Day-of-the-week effect Seasonaleffect
下载PDF
上一页 1 下一页 到第
使用帮助 返回顶部