...过直交脉冲矩阵来了解个别变量受本身与其它变量冲击的相对程度,因此可通过测量模式中不同变量的残差相关(Residual Correlation),来揭示个别变量之异常报酬率(Abnormal Return) ,并可从本身过去变化以及其它变量的信息传递来解释异常报酬率的比例。
基于16个网页-相关网页
利用信息熵的概念,文中建立了衡量残差相关性的度量函数。
Using the concept of information entropy, a measure function of the autocorrelation of residuals is established.
含噪语音经过线性预测分析后,所得的线性预测残差可分为与语音相关的语音残差部分和与噪声相关的噪声残差部分。
As the result of linear prediction analysis, the prediction residual of noisy speech can be divided into two parts: the speech residual related with speech and the noise residual related with noise.
该物理量融合了随机序列的相关性和信息论中信源熵的思想,综合反映了残差中所残留的信息量。
The quantity fuses correlation of stochastic process and idea of entropy in information theory, it indicates the amount of information left in residual series.
应用推荐