基于ABC和BP神经网络的铁路货运成本预测研究

运用作业成本法,结合BP神经网络理论建立铁路货运成本预测模型。通过案例分析,得出BP神经网络可以更准确的预测铁路货运成本费用的结论。     

支持向量机在模式识别中的应用

针对传统神经网络存在网络结构难于确定、过学习以及局部极小等问题，研究了基于支持向量机（SVM）的模式识别问题。通过对棋盘这种典型非线性二值问题的分类研究，分析了支持向量机的分类与泛化能力。支持向量机在分类和泛化能力方面远远优于传统神经网络。最后将支持向量机用于对两类飞机目标的分类识别，通过多组蒙特卡罗试验，获得了较好的识别结果。支持向量机在目标识别中有巨大潜力和广阔前景。     

我国企业跨国经营的战略选择

跨国公司是全球化背景下各国综合国力的重要标志，也是各国参与全球竞争的有力武器。加快发展跨国公司，努力缩小与发达国家的差距，是当前我国经济发展的重要任务，也是我国企业改革与发展的战略选择。本文剖析了我国企业跨国经营的现状与问题，提出了发展我国企业跨国经营的战略对策与建议。     

车辆自动识别系统移动站及其在城市交通监管中的应用

车辆自动识别系统 (AVI)移动站具有机动、快速和自动识别车辆的特点 ,适合于在城市监管中广泛应用。它采用射频识别 (RFID)技术 ,通过车卡 (TAG)采集车辆数据信息 ,对车辆进行自动识别和处理 ,识别率超过 99%;采用视频识别 (VFID)技术 ,通过摄像机采集车辆图像信息 ,在光照较强的情况下 ,识别率可以达到 80 %以上。     

利用ADS软件设计X频段低噪声放大器

首先简要介绍微波低噪声放大器的设计理论和方法,然后介绍使用Agilent公司的微波电路CAD软件ADS进行仿真和优化设计一个X频段的低噪声放大器的方法和过程。对制成品的实际测试和调试表明,此放大器达到了预定的技术指标,性能良好。其工作频段为8.6~9.5 GHz,噪声系数≤1.8 dB,增益为23 dB,带内平坦度≤±0.5 dB。     

中外物流业比较及我国的对策

在全面了解国内外物流业发展状况的基础上，提出了我国发展物流业的措施。     

民营银行设立路径选择与发展探讨

文章提出了民营银行应采取什么样的模式或路径发展才能发挥民营银行的优势,并阐述了民营银行的业务经营发展的思路,探讨了设立民营银行的必要性。     

Underwriter learning about unfamiliar firms: Evidence from the history of biotech IPOS

This study examines the ability of underwriters to properly value unfamiliar firms prior to issuance. I use a sample of IPOs in biotechnology, a relatively new but thriving industry. The first American biotech IPO was in 1980. Through the end of 2004, almost 500 biotech IPOs have appeared in the public market. I find that biotechnology differs from other industries in the attributes of individual firms valued by the market. In particular, R&D and the quality of human capital (e.g., star scientists on the staff) are much more important for biotech valuations. I find also that underwriters appeared not to appreciate this distinction for early biotech IPOs; in those cases, first-day market returns were predictable by firm attributes not used by underwriters to establish IPO issue prices. I also find that underwriters have learned over time, albeit slowly. Over the 20+ years of biotech history, IPO issue prices have become more dependent on firm attributes unique to biotechs while first-day market returns have become less predictable.     

Do alliances promote knowledge flows?

We explore the role of interfirm alliances as a mechanism for sharing technological knowledge. We argue that knowledge flows between alliance partners will be greater than flows between pairs of nonallied firms, and less than flows between units within single firms. Using patent citations as a proxy for knowledge flows, we find results that are consistent with these expectations. We then explore how firm characteristics affect knowledge flows within alliances and find positive effects due to technological, geographic, and business similarities between partners. We use alliance data from MERIT, patent data from the USPTO, and firm data from Compustat.     

昆明市生活垃圾处理与利用情况调查和评价

昆明市采用原生垃圾卫生填埋的单一方式处理垃圾,资源化程度低,同时对环境造成了严重的二次污染.昆明市生活垃圾处理应该采用基于垃圾分类(分选)的综合处理技术,即对于易降解的有机生活垃圾采用厌氧消化的处理方式,可以得到绿色能源和肥料;不易降解的部分进行回收或生产废弃物复合材料.卫生填埋只能作为垃圾产业化处理的最终处置方式.