从结论中可以看出其回归误差很小，满足一般工程的使用要求。
It concluded that the regression error is very little, and it satisfies the demand of the common engineering analysis.
对两条回弹曲线的回归误差进行对比分析，结果表明，考虑含水率的回弹曲线具有更高的检测精度。
Furthermore, regression errors of the rebound curves are compared and analyzed. The results show that rebound curves considering water content have higher test precision.
对应于方差参数这两种估计的回归系数的两种两步估计，它们的均方误差大致相当。
However, two-stage estimates of regression coefficients corresponding to these two estimates have approximate equal mean square error.
用幂函数模型描述的回归曲线以及根据试验参数得到模拟曲线与试验值比较误差均较小。
The errors of the data obtained using the power function model are very small from the test data.
回归计算中，由于资料和计算上的误差，使结果也产生一定的误差。
Due to the error of data and calculation, there are certain errors to the results in regression computation.
该模型由有限个回归类构成，每个类的模型误差可以是正态的，也可以是满足一定条件的任意分布。
The model consists of a finite number of regression classes, for each of them the model error may normal or arbitrary under certain conditions.
在消除回归模型误差项的自相关现象的处理中，提出了一种确定自相关参数估计值的新方法。
In this paper, a new process of estimating autocorrelation parameter is given when remedying autocorrelation of error term in regression model.
介绍了径向基函数网络的函数逼近原理和方法，提出了一种基于广义回归神经网络（GRNN）的传感器非线性误差校正方法。
The RBF network function approximation theory and method are introduced, and the method of nonlinear error correction of sensor is presented based on generalized regression neural network(GRNN).
在含有离群值的情况下，分析了最小二乘估计不能克服粗大误差对回归曲线的影响。
On the data including the discrete points, least square estimation can not control the strong error affecting the regression curve.
然后应用多元线性回归理论建立了关键测温点的温升和热误差的数学模型。
Then it established the mathematical model of temperature and thermal errors using the polynomial linear regression theory.
在此基础上建立时变序列预测公式及误差估计公式，给出其回归与时变自回归模型。
The prediction formula and its error estimation are also established. Its regression and time-varying autoregression model is presented.
在误差为相依的情况下，讨论了线性回归模型的刀切最小二乘估计与广义刀切最小二乘估计。
This paper studies linear regression models with dependent errors, and we introduce the jackknifed least squares estimator and generalized jackknife least squares estimator.
用幂函数模型描述的回归曲线以及根据试验参数得到模拟曲线与试验值比较误差均较小。
The errors are less as comparing the test data with regressive curve and simulative curve from test parameters using the power function model.
因此在第二章，本人用半参数方法提出了协变量测量有误差时的中值回归模型，用它来估计寿命变量的中值。
In the second chapter, I propose semi-parametric estimate to get median regression models with the measured error margin and use it to estimate the mediant of survival data.
而估计值的平均误差，却是度量观察值围绕着回归直线的变化程度或分散程度。
The standard error of estimate, on the other hand, measures the variability, or scatter, of the observed values around the regression line.
本文考虑在绝对误差和最小准则下的线性回归模型估计。
This paper considers the estimation in the linear regression model under the criterion of minimizing the sum of absolute errors.
对这类模型的统计建模，人们既关心回归系数的估计，更关心误差条件方差结构中未知参数的估计。
For this kind of modeling, people care for not only the estimation of regressive coefficient but also the estimation of unknown parameters in conditional skedasticity.
与其他建模方法进行比较可以看出，文中所建回归模型误差明显降低，满足工程精度需要，具有实际应用价值。
By comparing with other modeling methods, it is shown that Models in this paper have lower error ratio obviously, satisfy the engineering precision demand, and have application value.
线性回归模型的误差项不服从正态分布或存在多个离群点时，可以将残差秩次的某些函数作为权重引入估计模型来减少离群点的不良影响。
When multiple outliers occur in linear regression model or the distribution of residuals is not normal, we can use residuals rank as weight function to get some resist estimator.
对有限总体在不放回抽样的方法下，构造了一类非线性回归估计量，讨论了它的性质及其有关抽样误差．。
A kind of nonlinear regression estimator for limited Population is given and the quality and the sampling variance about is discussed.
本文分析了用脉冲法注入示踪剂时示踪剂的分布、时间与应答信号的测量，以及模型参数的回归等方面的种种误差。
The error in experimental determination of RTD derived from tracer pulse injection, digit time and response measurement, and flow model parameter estimation are discussed.
对回归方程误差的影响因素进行了分析，导出了回归方程的均方误差与其影响因素之间的关系式。
The influencing factors for errors in regression equation are discussed and the relations between mean square error of regression equation and its influence factors are deduced.
模拟的结果显示ANN模型比线性回归模型有更好的预测能力，预测的平均相对误差：ANN模型为14.9%，线性回归模型为25.8%。
Simulation results showed that the ANN model gave better predictions than the regressive model. The average relative error of ANN was 14.9% and that of linear regression was 25.8%.
提出了无偏回归估计方法，从理论上证明了该法比常规的估计法具有更小的平均估计误差。
It was proven theoretically that the estimate errors of unbiased regression estimation are less than that of general regression estimation.
回归分析时，应对回归方程的误差进行分析。
The error of regression equation should be analyzed when using regression analysis method.
将线性回归与时间序列分析进行比较，发现时序分析的平均绝对百分误差(MAPE)远远小于线性回归分析。
Comparing the linear regress analysis with the time series analysis, the latter's MAPE are less than the former.
并且组合模型进一步给出了回归分析中误差扰动项未来值的一些信息，提高了预测精度。
In addition, some information of the future value of random disturbance term in regression analysis is delivered by complex model, so that the forecasting accuracy is improved.
目的研究部分线性自回归模型中误差矩的估计。
Aim to study the estimates of error moments in partly linear autoregressive models.
目的研究部分线性自回归模型中误差矩的估计。
Aim to study the estimates of error moments in partly linear autoregressive models.
应用推荐