共查询到18条相似文献，搜索用时 921 毫秒
1.
为了改进用于分析大量影响因素的交通事故模型,采用基于马尔可夫链蒙特卡罗法和吉布斯抽样的条件自回归负二项模型来拟合过度散布性(由负二项过程拟合)、未观察异质性和空间相关性(由条件自回归过程拟合).统计检验显示,由于具有更小的预测误差和更强的参数估计,条件自回归负二项模型优于条件自回归泊松模型、负二项模型、零膨胀泊松模型和零膨胀负二项模型.研究结果表明,交通事故率和死亡人数与车道数、曲线长度、车道年平均日交通量和降雨量成正比.最大限速和最近医院距离与交通事故次数成反比,而与死亡事故次数成正比,这可能是由于过高的速度会引发更严重的事故以及救援伤者时丧失较长时间. 相似文献
2.
现有文化差异对中国企业对外直接投资（OFDI）区位选择影响研究较少考虑东道国情境因素的调节效应,结论也不一致。通过整合制度理论与资源基础观,研究东道国华人网络对文化差异与中国企业OFDI区位选择之间关系的调节效应。基于企业层面OFDI项目数据,采用面板负二项回归方法的实证研究发现,东道国与中国的文化差异对中国企业OFDI有显著的负向影响,但东道国华人网络削弱了文化差异的负向影响。在采用零膨胀负二项回归方法进行稳健性检验后,上述结论仍成立。根据实证研究结果对已有研究结论分歧进行了解释,提出了中国企业利用华人网络＂走出去＂、＂走进去＂以及更好地＂走下去＂的政策建议。 相似文献
3.
4.
研究的是保费收取的次数为负二项随机序列的复合负二项模型时的破产概率．对离散的经典风险模型进行改进，讨论了盈余的性质，给出了关于破产概率的一个定理，得到了破产概率的上限． 相似文献
5.
给出了负二项分布的两个不同定义,给出了两类负二项随机变量的期望、方差与矩母函数。从直观上对这两类负二项随机变量做了描述。 相似文献
6.
基于首都高校博士生调查数据，运用零膨胀负二项回归等模型分析“申请—考核”制对博士生选拔的效果。研究发现：“申请—考核”制选拔出的博士生专业基础更扎实、学术志趣更笃定，但与非“申请—考核”制博士生的科研能力无显著差别；“申请—考核”制的选拔效果不仅存在学科差异，还受到考试形式与选拔标准的影响。其中，完全取消笔试将削弱“申请—考核”制的选拔效果，而规定本科院校层次对“申请—考核”制的选拔效果不存在显著影响，设置论文发表数量的显著作用仅局限在博士生专业素养维度。基于此，建议从以下方面完善“申请—考核”制：注重构建科学有效的学科分类选拔标准体系，将笔试与面试有机结合，适度关注科研发表数量，谨慎设置本科院校层次条件。 相似文献
7.
线性混合模型是一类有着广泛应用的统计模型,对其中的方差分量,常使用方差分析法来估计。本文研究了在一种特殊情况下,含三个方差分量的线性混合随机效应模型的ANOVA估计,讨论了在何种条件下此估计在均方损失下一致优于ANOVA估计。由于此方差分析估计取负值的概率大于零,用在某非负点截尾的方法给出了方差分量的非负估计,并给出了得到的估计在均方损失意义下优于截尾之前的估计的充分条件。 相似文献
8.
一般地,正实数大于零,零大于负实数,两个正数绝对值大的数大,两个负实数绝对值大的反而小．两个无理数的大小比较较难,方法如下．一、根据被开方数的大小比较偶次方根大小,被开方数大的方根就大 相似文献
9.
《数学学习与研究(教研版)》2010,(1):25-26,I0003
1．如果两个有理数在数轴上的对应点在原点的同侧,那么这两个有理数的积（）．A．一定为正B．一定为负C．为零D．可能为正,也可能为负 相似文献
10.
梁艳 《宁夏师范学院学报》2008,29(6):6-11
本文针对LR-型模糊数,定义了一个指数型距离函数,在此基础上提出了一种稳健模糊线性回归方法．该方法将回归参数的估计转化为一个无约束最优化问题,其稳健性取决于指数型距离函数中一个参数的选择．数值试验表明,当在0．1与0．3之间取值时,所得模型稳健性较好;当趋近于零时,由该方法求得的模糊回归参数将退化为Diamond等人提出的最小二乘估计． 相似文献
11.
Christopher David Desjardins 《Journal of Experimental Education》2016,84(3):449-472
The purpose of this article is to develop a statistical model that best explains variability in the number of school days suspended. Number of school days suspended is a count variable that may be zero-inflated and overdispersed relative to a Poisson model. Four models were examined: Poisson, negative binomial, Poisson hurdle, and negative binomial hurdle. Additionally, the probability of a student being suspended for at least 1 day was modeled using a binomial logistic regression model. Of the count models considered, the negative binomial hurdle model had the best fit. Modeling the probability of a student being suspended for at least 1 day using a binomial logistic regression model with interactions fit both the training and test data and had adequate fit. Findings here suggest that both the negative binomial hurdle and the binomial logistic regression models should be considered when modeling school suspensions. 相似文献
12.
The study tests 3 research hypotheses on the teaching-research nexus using the administrative data of the HSE University in Russia for 7 years. We confirmed the hypothesis that a lower classroom teaching load is associated with higher research output. Undergraduate courses have a relatively higher negative effect, especially on the probability of producing research output at all. A higher concentration of teaching in some semesters or quarters looks rather ineffective as a way to stimulate higher research productivity. Only a moderate share of English-taught classes (about 20% of the teaching load) is associated with higher research output and only for publications in good international journals. The empirical part of the study is based on zero-inflated Poisson and zero-inflated negative binomial models. This study comes to contribute to the development of literature discussing a mechanism and possible implications in teaching load composition as an essential part of the academic contract. 相似文献
13.
Hui Liu 《Structural equation modeling》2013,20(2):247-279
This article applies growth curve models to longitudinal count data characterized by an excess of zero counts. We discuss a zero-inflated Poisson regression model for longitudinal data in which the impact of covariates on the initial counts and the rate of change in counts over time is the focus of inference. Basic growth curve models using a single outcome are described, as well as a model in which two linked outcomes constitute a dual-trajectory growth process. This model is applied to assess the impact of changes in financial stress on longitudinal change in smoking. 相似文献
14.
基于2020年“全国研究生满意度调查”13264份学术学位博士研究生的数据,运用线性回归与标准负二项回归分析了不同招生录取方式的博士生在主观教育收获满意度以及客观论文发表数量方面的差异。分析结果显示,本科直博与硕博连读的博士研究生与公开招考的博士生相比,其论文发表数量更多,但博士教育收获满意度更低。建议博士生培养单位充分发挥贯通式人才培养连续性、系统性的优势,合理设计长学制博士生课程教学、导师指导以及考核管理制度。 相似文献
15.
An Investigation of Classification Consistency Indexes Estimated Under Alternative Strong True Score Models 总被引：1，自引：0，他引：1
Using several data sets, the authors examine the relative performance of the beta binomial model and two other more general strong true score models in estimating several indexes of classification consistency. It is shown that the beta binomial model can provide inadequate fits to raw score distributions compared to more general models. This lack of fit is reflected in differences in decision consistency indexes computed using the beta binomial model and the other models. It is recommended that the adequacy of a model in fitting the data be assessed before the model is used to estimate decision consistency indexes. When the beta binomial model does not fit the data, the more general models discussed here may provide an adequate fit and, in such cases, would be more appropriate for computing decision consistency indexes. 相似文献
16.
We find the correlation of two jointly distributed random variables connected with a coin tossing experiment. The marginal distributions are binomial and negative binomial. 相似文献
17.
《Child abuse & neglect》2014,38(10):1569-1580
The purpose of this study was to evaluate the interrelationship among childhood abuse and traumatic loss, posttraumatic stress symptoms (PTSS), and Axis I psychiatric disorders other than PTSD among newly incarcerated adults, and to test a proposed model in which the severity of PTSS mediates the relationship between childhood abuse/loss and adult psychiatric disorders. Four hundred sixty-five male and female inmates participated in a structured clinical research interview. Four types of interpersonal potentially traumatic experiences (physical abuse, sexual abuse, emotional abuse, and traumatic loss) were assessed for occurrence prior to the age of 18 years old. Current psychiatric disorders and PTSS were also assessed by structured interview. Negative binomial regression was used to evaluate the association between the cumulative number of types of childhood abuse/loss experienced and number of current Axis I disorders, and to test the mediation model. Approximately half of the sample (51%) experienced 1 or more types of childhood abuse/loss, and 30% of the sample had at least one psychiatric disorder other than PTSD. For both men and women, childhood physical abuse and childhood sexual abuse were independently associated with psychiatric morbidity, and an increasing number of types of childhood trauma experienced was associated with an increase in the number of current Axis I diagnoses. However, these associations were no longer statistically significant when severity of PTSS was added to the model, providing support for the proposed mediation model. Implications for secondary prevention services for at-risk inmates are discussed. 相似文献
18.
二项分布可靠度的Bayes估计 总被引：3，自引：0，他引：3
周燕燕 《西安文理学院学报》2008,11(4):50-53
在二项分布的可靠度的先验分布为幂分布时给出了其在熵损失函数下的E—Bayes估计公式和多层Bayes估计公式．通过实例验证,这两种估计公式都是稳定的,并进一步表明了在熵损失函数下计算出的二项分布可靠度的E—Bayes估计比多层Bayes估计的精度更高,稳定性更好,计算简单,便于应用． 相似文献