一种快速时变自适应信道估计盲方法

提出了一种新的信道估计最小均方(LMS)盲方法--时变步长理想加权判决信道估计盲方法.该方法通过对常规LMS算法进行改进,实现自适应跟踪信道特征变化;同时利用接收机判决误差信息函数作为权系数的理想加权判决,从而解决了常规LMS盲方法收敛速度慢、信道估计性能不高的缺点.仿真证明,对于不同的多普勒频移、时延扩展以及时间,该方法均表现出了比常规LMS方法更优的性能.同时,该方法还可用于航天、雷达等领域估计其他特征参数.     

中美贸易顺差与人民币汇率关系的实证分析

针对中美贸易顺差与人民币汇率关系的问题,本文通过协整分析和Granger因果检验发现,人民币升值在短期难以对中美贸易顺差产生调节作用,从长期影响而言,人民币升值的影响作用也不大。此外,中美贸易顺差在短期还是主要取决于当前贸易收支的现实情况,而中长期则在很大程度上受到美国经济波动的影响。因此,人民币升值不会缓解美国巨大的贸易逆差,而且即使美国贸易收支出现改善也不会对美国经济发展产生显著的正面影响。     

企业规模对专利产出影响的实证分析——基于1998—2010年上市公司经验证据

使用1998—2010年中国A股上市公司的年度数据,研究了企业规模对专利产出和专利密度的影响。结果表明:发明专利和实用新型专利与企业规模之间存在U型关系,外观设计专利与公司规模呈倒U型关系;三类专利的专利密度都与企业规模之间存在U型关系;企业的知识存量、成立时间、财务杠杆、所有制类型等企业内部因素以及市场集中度、企业所处区域等外部环境因素都对企业的专利产出有显著影响。据此提出企业创新发展路径假说。     

Measuring firm performance: Differentiating between uncontrollable and controllable bad outputs

This paper introduces a new model that differentiates between controllable and uncontrollable bad outputs in the measurement of firm performance; the airline industry is used as the empirical example. The model allows us to measure technical inefficiency, as well as the inefficiency in the production of bad outputs, and the inefficiency in the investments allocated to the reduction of uncontrollable bad outputs. Moreover, we provide a direct measure of the total cost of controllable bad outputs. As the model involves a number of latent (unobservable) variables that cannot be easily integrated out of the likelihood function without producing complicated expressions, we resort to Bayesian techniques. We find notable differences in inefficiency across airlines. More importantly though, we find that the production of bad or undesirable outputs impose significant costs on airlines. However, these cost issues could be reduced if airlines invest more in maintenance and service training.     

Selection of influential variables in ordinal data with preponderance of zeros

Presence of excess zero in ordinal data is pervasive in areas like medical and social sciences. Unfortunately, analysis of such kind of data has so far hardly been looked into, perhaps for the reason that the underlying model that fits such data, is not a generalized linear model. Obviously some methodological developments and intensive computations are required. The current investigation is concerned with the selection of variables in such models. In many occasions where the number of predictors is quite large and some of them are not useful, the maximum likelihood approach is not the automatic choice. As, apart from the messy calculations involved, this approach fails to provide efficient estimates of the underlying parameters. The proposed penalized approach includes ?₁ penalty (LASSO) and the mixture of ?₁ and ?₂ penalties (elastic net). We propose a coordinate descent algorithm to fit a wide class of ordinal regression models and select useful variables appearing in both the ordinal regression and the logistic regression based mixing component. A rigorous discussion on the selection of predictors has been made through a simulation study. The proposed method is illustrated by analyzing the severity of driver injury from Michigan upper peninsula road accidents.     

Arrow–Debreu equilibria for rank‐dependent utilities with heterogeneous probability weighting

We study Arrow–Debreu equilibria for a one‐period‐two‐date pure exchange economy with rank‐dependent utility agents having heterogeneous probability weighting and outcome utility functions. In particular, we allow the economy to have a mix of expected utility agents and rank‐dependent utility ones, with nonconvex probability weighting functions. The standard approach for convex economy equilibria fails due to the incompatibility with second‐order stochastic dominance. The representative agent approach devised in Xia and Zhou (2016) does not work either due to the heterogeneity of the weighting functions. We overcome these difficulties by considering the comonotone allocations, on which the rank‐dependent utilities become concave. Accordingly, we introduce the notion of comonotone Pareto optima, and derive their characterizing conditions. With the aid of the auxiliary problem of price equilibria with transfers, we provide a sufficient condition in terms of the model primitives under which an Arrow–Debreu equilibrium exists, along with the explicit expression of the state‐price density in equilibrium. This new, general sufficient condition distinguishes the paper from previous related studies with homogeneous and/or convex probability weightings.     

Is DNA fingerprinting the gold standard for estimation of adoption and impacts of improved lentil varieties?

In the early 1980s, disease susceptibility in short-season lentil landraces began to limit productivity in areas where relay cropping took place in Bangladesh. Since then, several improved high-yielding lentil varieties, which are resistant to rust and blight and suitable in the relay cropping system, have been released jointly by national and international research centers. This study used three methods, namely a panel of experts, a survey of 1000 households where the respondents named the variety they used, and DNA fingerprinting of seed samples collected from all lentil plots cultivated by survey households to estimate adoption. Double hurdle and instrumental variables regression methods were applied to the household survey and DNA fingerprinting data to identify determinants of adoption and measure their impacts. Of particular interest was whether estimates of adoption, determinants of adoption and impacts varied by method of variety identification. Results showed that the expert panel overestimated the adoption of more recent varieties while about 89% of the farmer-reported varieties were accurate, as verified by DNA fingerprinting. DNA fingerprinting appears to have little advantage for estimating the level of adoption in this case, where few varieties of lentils are found, local variety names do not exist, and most seed is obtained through a formal system. However, even under these conditions, determinants of adoption vary by identification method, and use of farmer-reported information on the variety can lead to erroneous conclusions about determinants of adoption. Because recent breeding efforts have focused on taste and cooking considerations, yield impacts were not significantly different from zero.     

一种基于自适应相似矩阵的谱聚类算法

为了消除在构建谱聚类算法的相似矩阵时,高斯核函数中尺度参数的波动影响,构建了一种自适应相似矩阵,并应用到谱聚类算法中。自适应相似矩阵中数据点间的距离度量采用测地距离算法,相距较近的两点间的距离近似于欧氏距离,相距较远的两点则先根据欧氏距离得到每个数据点的k个近邻点,然后累加近邻点的测地距离,由此得到每对数据点间的最短距离。两点间的局部密度用共享近邻的定义来表示,更好地刻画了数据集的本征结构。在5个人工数据集和国际通用UCI数据库中的5个真实数据集上进行实验。实验结果表明,所提算法的聚类准确率高于对比算法的准确率,对复杂分布数据有很强的自适应能力。研究成果为数据挖掘及机器学习提供了思路和方法。     

Some Variants of Constrained Estimation in Finite Population Sampling

This is a review article that unifies several important examples using constrained optimisation techniques. The basic tools are three simple mathematical optimisation results subject to certain constraints. Applications include calibration, benchmarking in small area estimation and imputation. A final illustration is constrained optimisation under a general divergence loss.     

重叠滤波多音系统的频偏性能分析及优化

重叠滤波多音(O-FMT)是针对滤波多音(FMT)系统频谱利用率低的缺点而根据超奈奎斯特(FTN)概念引入子载波重叠得到的方案。鉴于O-FMT系统在频偏环境下的性能缺陷，理论分析了频偏对系统信号各部分的影响，比较了O-FMT与正交频分复用（OFDM）系统的抗频偏性能。针对频偏性能优化问题，提出了基于最佳线性无偏估计（BLUE）的改进盲估计算法，设计了基于频域滤波器系数的内嵌结构进行频偏补偿。仿真实验表明，O-FMT相比OFDM具有更好的抗频偏能力，改进的盲估计算法估计值精度更高，抗干扰、抗衰落能力更强，设计的频偏补偿结构可以使系统获得更好的性能。