In the technique of simulation, the theory of conditional expectation and conditional distribution i...
而由此引起的对后续状态转移过程的影响，是通过一种递推条件分布来考虑的。
Therefore, the analytical model has allowed for the distribution of operating expenses to be conditional upon the level of gross receipts realized in any year.
因此，分析模型，使得空间分布的营业费用是无条件的总收益水平在任何一年。
For the distribution of the flowfield in the flow measurement by ultrasonic, the measuring precision is improved by the circuit of ultrasonic emission in using high voltage and narrow tripping pulse, the circuit of high frequency oscillation and counter, and the circuit of phase demodulation. The result is compensated by the conditional data that has...
针对超声检测流量中的流场分布情况，采用高电压窄脉冲信号触发超声波发射电路、高频振荡计数与相敏检波相结合的高精度在线检测方法提高测量的精度，并利用系统内存储的测量环境数据和实际测量时的温度对测量结果进行补偿，保证测量的稳定性；分析了测量误差来源以及消除误差的方法。
As the researchs on its distribution are introduced in numerous literatures, its research methods are to work out its single distribution first and then the joint distribution, or to make conjectures upon first and prove the joint distribution of multiple order statistics by inductive method, and then study the distribution of single order statistics by the results, and finally make a system research on its conditional distribution.
或先猜想然后用归纳法证明多个顺序统计量的联合分布，接着根据联合分布研究单个顺序统计量的分布，最后对其条件分布做系统研究。
Solid powder was adopted as unconditional seeding to study the flow characteristics of global field and atomized silicon oil droplets were used as conditional seeding to study the flow characteristics of reactants only. The measurement result of unconditional seeding, which does not distinguish the velocity of reactants or that of products, shows that in reacting zone, the measured velocity had a probability density function of two-peak distribution. The result tentatively demonstrates that in combusiton zone, the laser probe was alternatively subjected to low-speed reactants and high-speed products, due to flame flickering. The compounded measured velocities enlarges the span of velocity PDF and leads to a high turbulence. The measurement with conditional seeding, which limits the result to only those of reactants, shows no apparent difference between in flame zone and far upstream zone, in terms of mean velocity, turbulence and velocity PDF. The result affirmed the conclusion that the peak-turbulence in reacting zone is due to flame flikering.
采用雾化硅油只对反应物流场进行示踪，结果表明，脉动的、皱褶的预混火焰锋面对反应物流场性质并没有大的改变，在反应区内来流仍然保持原有的管流特征；采用耐高温固体粉末对全流场进行示踪，结果表明在火焰区内速度的概率密度分布为双峰分布，其中速度较低的峰的中心速度与硅油示踪得到的平均速度一致；火焰区湍流度的增大主要是由于测量探头交替地处于高速的产物和低速的反应物之间，导致速度测值分散所致，得到的仅仅是统计意义上的结果，为表观湍流度。
Monte Carlo is a method that approximately solves mathematic or physical problems by statistical sampling theory. When comes to Bayesian classification, it firstly gets the conditional probability distribution of the unlabelled classes based on the known prior probability. Then, it uses some kind of sampler to get the stochastic data that satisfy the distribution as noted just before one by one. At last, it can obtain the posterior probability distibution of each unlabelled classes by analysing these stochastic data.
蒙特卡罗是一种采用统计抽样理论近似求解数学或物理问题的方法，它在用于解决贝叶斯分类时，首先根据已知的先验概率获得各个类标号未知类的条件概率分布，然后利用某种抽样器，分别得到满足这些条件分布的随机数据，最后统计这些随机数据，就可以得到各个类标号未知类的后验概率分布。
This regression problem is dicult for several reasons: large number of examples, large number variables (most of which are discrete and multi-valued), non-stationarity of the distribution, and a conditional distribution of the dependent variable which is very dierent from those usually encountered in typical applications of machine learning and function approximation.
这个回归问题是，邸邪教组织有多少个原因：大量的例子，大批变量（其中大部分是离散和多值），非平稳性的分布，以及有条件的分布因变量，这是非常邸erent从这些一般遇到的典型应用机器学习和函数逼近。
Firstly, the Signal Noise Ratio characteristic of matched filter despread symbol and the distribution of coherent phase estimation error are derived. Secondly, the exact integration formula of BER for fading channel is presented, moreover, by using the Maclaurin series expansion of MPSK conditional Symbol Error Rate, the simplified BER approximations for BPSK and QPSK are derived over AWGN, Nakagami-m fading and Rayleigh fading channel.
首先推导了匹配滤波解扩符号的信噪比特性和相干相位估计误差的分布；然后给出了衰落信道中误比特率的精确积分式，并利用MPSK条件误符号率的麦克劳林级数形式，推导了AWGN，Nakagami-m衰落和瑞利衰落信道中BPSK，QPSK误比特率的简化逼近式。
In addition, we show that the transformed conditional distribution function is of the same family as that of the original one as long as the moment generating function of the original distribution function exists, which is illustrated by two special cases:Normal distribution and Normal Inverse Gaussiandistribution.
此外，本文还通过两个特殊情况(正态分布和正态逆高斯分布)说明了：只要初始分布函数的母函数存在，转化后的条件分布函数保持原有的分布类型。
Traditional methods of modelling returns and testing the Capital Asset Pricing Model do so at the mean of the conditional distribution.
传统的模拟与检验资本资产定价模型(Capital Asset Pricing Model，CAPM)及各种资产定价模型的方法都是检验模型在均值处的有效性，但是实证结果不同，带来了有很大的争议。
Aside from the ordinary least square evidence in the existing studies, this thesis employs quantile regression model to tests selection hypothesis and productivity hypothesis in consequence by considering selection of marriage, wife`s labor force participation and labor market hours in the models, focusing on the possible heterogeneity of male marriage premium over the conditional distribution of wage.
有别於现有文献大多在普通最小平方法架构下探讨婚姻溢酬，本研究针对生产力假说与选择性假说，考量婚姻选择性、妻子劳动参与选择性以及工时内生性对男性工资率的影响，采用分量回归模型观察台湾男性婚姻溢酬在工资条件分配的异质性。
Assuming that innovation follows the conditional t distribution and conditional Generalized Error Distribution.
假设新息服从t分布和广义误差分布两种条件分布。
In statistics, a statistic is sufficient for the parameter θ, which indexes the distribution family of the data, precisely when the data's conditional probability distribution, given the statistic's value, no longer depends on θ.
统计量是样本的不带任何未知量的函数，一般而言，统计量所包含的信息比样本要少，但可能这些漏掉的信息是无关紧要的。
The shortest confidence interval of parameter of Gamma distribution is in practice a question of conditional extremum. It can be transferred to a group of equations that can be solved by a method of numerical value.
Γ分布中未知参数θ的最短置信区间实际上是一个条件极值问题，它能转化成一个方程组可用数值方法迭代求解。
A multi-objective estimation of distribution algorithm is proposed. This algorithm uses decision-tree-based Bayesian network as probabilistic model to encode conditional dependencies among variables.
提出一种采用贝叶斯网络表示概率模型的多目标分布估计算法(multi-objective estimation of distribution algorithm，MEDA)。
Methods for feature space discretization based cluster algorithm and class conditional probability distribution estimation are proposed.
首先针对应用贝叶斯决策理论时难以估计类条件概率密度函数的问题，提出了特征空间离散化及相应的类条件概率分布的估计方法。
For CAPM beta, there is no significant risk premium by OLS, but when using the technique of QR, we can detect that the market price of CAPM beta is significant in both tails of the conditional distribution – negative for the firms that underperform and positive for the firms that overperform.
在各风险因子对於横断面股市报酬之解释力上，以CAPM贝它值而言，在未区分市场投资组合超额报酬率时，普通最小平方（Ordinary Least Square，OLS）均值估计结果发现该风险因子对横断面股市报酬是不具有显著解释力；然而进一步以分量回归检视，却发对於相对高报酬率股票，其风险与报酬间为一显著正向关系。
Generally speaking, the Bayes rule provides us a suitable way to combine those two kinds of knowledge so as to generate a conditional distribution. By applying the Maximum Likelihood Estimation method to this distribution, the object's accurate positions could be estimated.
利用贝叶斯理论构造了一个条件概率分布，将以上两种模型相关参数归并起来，并在这个分布上应用最大似然佑计方法来估测目标的位置。
Bayesian network is a directed acyclic graph, it can use a conditional probability distribution directly express the dependencies between variables.
贝叶斯网络是一个有向无环图，能够通过一个条件概率分布直观地表达各个变量之间的依赖关系。
Key Distribution and Subscriber Addressing Optimization for Conditional Access System
条件接收系统中的密钥分配及寻址优化研究