时间序列分析方法及人民币汇率预测的应用研究

在简要介绍时间序列模型的基础上，使用人民币／美元的日汇率值进行实证研究，建立相应的ARIMA模型和EGARCH模型并进行预测和评价。研究结果表明，EGARCH模型的预测结果较ARIMA模型理想，适合描述人民币／美元汇率的变动趋势。     

改进实孔径雷达的方位分辨力

实孔径毫米波FMCW成像雷达系统中方位分辨力与距离分辨力不匹配,影响了成像质量。本文从分析“波束效应”的本质原因出发,提出了一种利用卷积反演—广义逆滤波法,对波束进行“伪压缩”是有效地提高成像雷达方位分辨力的方法。文中给出了对雷达实测数据作“伪压缩”处理的结果,可见处理后的方位分辨力提高约4倍。     

猪与人类基因比较定位研究的初步探讨

本文在综合猪与人基因定位研究最新进展的基础上,开展比较定位研究。由此对尚未定论的猪TPIl、PGMl和PEPB等基因的染色体定位提出了自己的见解,并分析确定人基因组中存在猪HAL连锁群和SLA连锁群的保守的同源染色体片段,这对深入研究此二连锁群有一定指导意义。本文还讨论了上述研究的重要意义。     

叶轮机械跨音速流动计算方法评述

本文从工程应用的角度，对现有的叶轮机械内跨音速无粘流动的计算方法进行了综合评述．由于激波装配法在叶轮机内的应用范围较窄，本文主要评述应用范围广泛的激波捕获法，以期对这一计算领域有一个概括性的了解．     

带有线性约束的多元回归模型中回归系数的线性估计的可容许性

针对矩阵风险，给出了矩阵估计量的多种优良性准则．在这些准则下，得到了带有线性约束的多元回归系数的线性估计是可容许的充要条件，进而由于不同的线性约束所引起的可容许的估计类的不同，我们得到了不同的估计类之间的一种刻划。     

A statistical analysis of the life expectancy of the residents of our country, 1973-75]

Statistical analysis of life expectancy is important in assessing population health and its characteristics and in studying human diseases and natural population changes. Life tables are constructed and statistical analysis is performed retrospectively on data accumulated over a 3-year (1973-1975) period. The data were supplied by the Office of Cancer Prevention and Treatment of the Ministry of Health, which originated from 24 provinces, representing an accumulated population of 2.04 billion, with a total mortality of 15.29 million. Results show that life expectancy in China has greatly improved since Liberation. Thus, in 1935, the average life expectancy for Nanjing residents was under 35 years. In 1951, the average life expectancy for male and female residents of Shanghai were 42.74 and 46.76 years respectively. But for the 1973-1975 period, the average Chinese life expectancy was 63.62 years for males and 66.31 years for females, with higher life expectancy for coastal provinces than for inland provinces. Cardiovascular diseases (excluding arteriosclerotic heart diseases), malignant tumors, and cerebrovascular diseases were the major causes of death in regions with higher life expectancy, while respiratory diseases, infectious diseases, and diseases of the newborn were the major causes of death in regions with lower life expectancy.     

我国区域经济增长收敛的计量分析——基于面板数据分析

本文综合运用回归分析和时间序列分析法,从不同角度较全面地考察了我国区域经济增长收敛问题。计量结果表明:改革开放以来,我国各省区的经济增长不存在收敛趋势,但东、中、西部三大经济带内基本上可以认为存在具有不同稳定性的收敛俱乐部现象,即东部经济带存在稳定的收敛趋势,任何外部冲击都应该是暂时的,而中、西部经济带却经常会由于外部冲击而随时可能呈现发散。     

Sample sizes for the transmission disequilibrium tests: tdt,s-tdt and 1-tdt

The transmission/disequilibrium test (TDT) is widely used to detect the linkage disequilibrium between a candidate locus (a marker) and a disease locus. The TDT is a family-based design and has the advantage that it is a valid test when population stratification exist. The TDT requires the marker genotypes of affected individuals and their parents. For diseases with late age of onset, it is difficult or impossible to obtain the marker genotype of the parents. Therefore, when both parents marker genotypes are unavailable, Ewex and Spielman extended the TDT to the S-TDT for use in sibships with at least one affected individual and one unaffected individual. When only one of the parents' genotype is available. Sun et al. proposed a test the 1-TDT, for use with niarker genotypes of affected individuals and only one available parent. Here, we study the saniple sizes of TDT, S-TDT, and 1-TDT. We show that the sample size needed for the 1-TDT is rogghly the same as the sample size needed for the S-TDT with two sibs and is about twice the sample size for the TDT.     

拓展基尼系数及其应用的拓展研究

本文探讨适合组数据的拓展基尼系数计算公式,给出了协方差、回归系数等算法,论证了拓展基尼系数满足的若干公理性质;纠正了现有拓展集中度指数计算中存在的问题,重新定义了拓展健康不平等测度和健康绩效指数;给出税收拓展累进性指数,并利用税收边际效应定义了一个新的税收累进性测度。同时使用以上方法,对我国各地区孕产妇保健的公平性、地方财政收入的税收累进性进行了实证分析。