参数估计

The parameter re-estimation algorithms of MDHMM and MCHMM are also discussed, and relatively simulations of the two models are made, in theories and simulations which present the accuracy and superiority of those HMMs.

还讨论了两种新模型MDHMM和MCHMM的参数重估计算法，并进行了相应的模拟实验，从理论和实验结果论证了新模型的精确性和优越性。

Connecting the theoretical analysis with the experimental results, reveals that the theory established here not only can describe the radiation induced trap law well, but also can evaluate the relation between the radiation life and the relevant parameters.

结合实验结果可以看到，所建理论不仅可以很好地描述实验结果的规律性，而且可以估计出辐照作用下介质的寿命（也即介质耐辐照所能吸收的极限剂量）及其与有关参数之间的联系。

In this text, we discuss the method of estimating parameter that the density of explains variable of having measurement error is not abridge, we inquire the covariance of y and e^ and we can get a formula as follows

本文要探讨的是在不限制有测量误差的解释变数的分布下估计参数的方法，首先探讨Y跟e^的共变异数关系，可以得到如下式子

Based on the tomato and eggplant supply data of Shanghai in sequence of year, using meteorological data in SAMR(The data management system of Shanghai Agricultural Meteorology Report) and screening the meteorological factors related to the vagetable output, the authors carried out curve fitting and estimation of related parameters of distribution function.

Gompertz曲线和对数正态分布函数(Logarithmic normal distribution function)作为这二种蔬菜上市趋势的数学模型，进行曲线拟合，先调用上海市农业科学院测试中心的SAMR农业气象资料数据库所存贮的气象数据，筛选与该蔬菜产量有密切相关的气象因子，再用多元回归方法对初选因子作进一步的筛选，从而估计出分布函数的有关参数。

The main results are:(1) The cumulate cause-effects model was demonstrated at both macro and micro level data. The problem of data fitting of hypertension prevalence data was solved well enough with the cumulate cause-effects model. The parameters of the model were meaningful for the practice.(2) Taking the hypertension developing speed as an indicator of health equality of hypertension between urban and rural residents in China and discovered the macro level determinants of it.(3) The causal network was testified by the structure equation model.(4) It is found that the major determinants of hypertension prevalence in China is population ageing, overweight, lifestyle changing due to high economic developing speed and the lagging behind of health education.

主要的研究成果有：(1)首次观测到我国高血压的患病率曲线符合累积风险效应模型，解决了高血压患病率资料数据拟合中长期存在的方法学问题，并对模型中参数的实际流行病学意义给出了合理的解释，能够对疾病的发病率进行间接的估计；(2)提出了用高血压的发展速度作为衡量我国城乡高血压造成的健康不平等的重要指标，解决了我国高血压患病率评价中长期存在的一些错误性认识，揭示了我国高血压城乡发展不平衡的宏观决定因素；(3)利用结构方程模型对构建的网络模型进行拟合分析，证明了模型的有效性；(4)揭示了目前影响我国高血压发展趋势的决定因素，主要包括人口老龄化趋势加剧、超重人口比例增加、经济的快速发展带来生活方式的转变和健康教育的相对滞后等几个方面。

According to Iacoviello and Minetti (2003), this article uses Taiwan data for 1995 first season to 2004 fourth season to check wheter Taiwanese housing market has the credit channel. This article establishes four groups of VAR model. Model (1) including overnight interest rate, GDP, the consumer price index, the house price index, the total bank loans, the bank housing loan. The model (2) replaces bank loan and housing loan; for spread. The model (3) uses Mix to replace the Spread. The model (4) including GDP, the consumer price index, the house price index and uses Mix to replaces overnight interest rate.

摘要 本文用台湾地区1995第一季至2004第四季共40笔季资，验证台湾房屋贷款市场是否存在信用管道，本文首先以ADF单根检定法检定各项变是否为定态数列资料，其次将非定态变数取差分值之后进行VAR模型分析，最后经由估计后VAR模型之参数进行冲击反应分析与预测误差变异数分解，藉以了解当变遇到外冲击时对於台湾房屋贷款市场的影响。

The nonparametric kernel density theory and methods are introduced to model the diameter distribution of broadleaved Pinus koraiensis in Liaoning eastern mountains in contrast with the normal distribution and Weibull distribution.

应用非参数核密度估计理论和方法，对位于辽宁东部山区的阔叶红松林标准地的红松进行直径结构模拟，并与正态分布、威布尔分布进行了比较。

Finally, in view of insufficient crosssectional data, pooled regression is used to analyze the relationshipbetween banking efficiency and corporate governance.

最后，考虑到解释变量多而横截面数据少，使用混合回归来估计参数。

The experiments proved that this algorithm can preferably restore the optical slices and estimate the parameters of the PSF.

经过实验验证，该算法能够较好地复原光学切片的图像，并且估计出PSF的参数。

The high resolution joint estimation method of multidimensional parameters of coherent signals with stereoscopic array.

采用立体阵的相干信号多维参数高分辨联合估计方法。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。