为了诊断具有异方差的线性回归模型的异常点，建立了具有异方差的均值漂移模型和数据删除模型。
The data deletion model and the mean drift model are equivalent in a general linear regression model.
其次，在误差服从正态分布的条件下，阐述了线性回归模型框架中均值漂移模型与异常点检验的等价性。
Secondly, we show the equivalence theorems between mean shift model and outlier detection model based on the condition that errors in regression model are normally distributed.
此模型的漂移项为非线性形式，具有显著的均值回复效应，且利率波动对利率水平极为敏感。
This model has nonlinear drift, has significant mean-reverting effect, and the volatility is highly sensitive to interest level.
跟踪算法分为两步进行，首先应用均值漂移算法实现手部区域的定位，然后基于自适应活动轮廓模型提取手部轮廓。
The proposed method consists of two steps: hand location using mean shift and hand extraction based on adaptive active contour model.
第五章讨论了均值漂移对带约束线性模型条件岭估计的影响问题。
The fifth chapter discusses the influence of mean shift disturbance to conditional ridge estimation in restricted linear model.
我们首次研究了方差加权和均值漂移的混合模型，得到了相应的诊断统计量。
We study the diagnostic model with case-weights and mean-shift simultaneously, and some new diagnostic statistics are derived.
研究了带有约束的均值漂移和方差加权的混合非线性回归模型．得到了相应的一阶和二阶诊断统计量。
We study restriction of Nonlinear Regression Diagnostic Models with case-weights and meanshifte simultanously, and some new diagnostic statistics are derved.
本文在对连续自适应均值漂移算法深入分析的基础上，提出利用目标外观信息的先验知识，对其建立多个颜色分布模型。
In this paper, we research the CamShift tracker in detail and then proposed to build multiple color-distribution-model for the target accordng to prior knowledge of the objects appearance.
实验结果显示，均值漂移算法分割结果对其参数变化较为敏感，而基于图模型的算法和混合算法则较为稳定。
Experiments' results show that the mean shift based algorithm is more sensitive than the graph based algorithm and the hybrid algorithm, which are very stable to parameters and different images.
实验结果显示，均值漂移算法分割结果对其参数变化较为敏感，而基于图模型的算法和混合算法则较为稳定。
Experiments' results show that the mean shift based algorithm is more sensitive than the graph based algorithm and the hybrid algorithm, which are very stable to parameters and different images.
应用推荐