具有AR(p)误差的非线性模型异方差和相关性检验

讨论随机误差是AR(p)序列的非线性回归模型的异方差和自相关性检验问题.首先导出联合检验的score统计量,然后利用参数的正交变换,得到了调整的score检验统计量.当模型存在自相关性时,给出了检验异方差性的score统计量和调整的score统计量.最后利用得到的检验方法分析了氯化物数据,分析结果表明,该数据具有显著的异方差和AR(2)相关性.     

基于多源数据的城市活力区域识别及驱动因素评估——以上海市为例

活力是评价城市空间品质的重要指标,也是城市空间品质提升的目标之一.城市是一个复杂的巨系统,城市的活力也因此受到多元、复杂的各种因素的影响.聚焦于街道/乡镇这一基本行政空间单元,运用ESDA(exploratory spatial data analy-sis)方法,结合信息熵模型,对上海市城市活力区域进行了识别评估,对...     

Using the bootstrap for improved ARIMA model identification

This paper presents a new method of identifying ARIMA time-series models. We use the bootstrap technique in estimating the distribution of sample autocorrelations both separately and in a simultaneous inference setting. The bootstrap has the advantage of being nonparametric and thus free of reliance on asymptotic normality, which may not hold for short or medium-size series. The simultaneous procedure is unique, as it has no feasible parametric alternatives. An application to exchange rates illustrates our methodology. In the example chosen, we are able to produce better forecasts using the model identified via the bootstrap technique.     

基于神经网络的混沌系统状态预测

在对混沌时间序列的未来状态进行预测时,必须能够正确确定重构序列相空间时的最优时滞及最优嵌入维数,并根据序列状态的变化选定适当的模型进行预测。利用时间序列的自相关函数来确定时滞τ值,利用混沌序列嵌入维数与关联维数的关系来确定最优嵌入维数m值。在此基础上,选择神经网络模型来逼近真实系统,并采用一种新的算法来确定网络权重。最后,通过Logistic映射所产生的时间序列对所述理论进行了实证研究。     

上证股市的弱型有效性及其演进

在简要评述弱型有效性检验方法的基础上,以1990-12-19～2002-06-21的上证综合指数为对象,运用6种典型的检验方法,包括两种形式的游程检验、自相关和Q检验、同方差以及在国内当前研究中尚未发现其应用的异方差条件下的方差比率检验,分别考察日收益和周收益序列整体样本和分段子样本,并简要探讨实证结论的原因.     

夜间灯光数据下河北省能源消费量的空间模拟

以NPP-VIIRS夜间灯光为数据源,对河北省各地级市能源消费数据与夜间灯光数据进行相关性分析,确定相关系数最高的二次多项式模型作为能源消费的回归模型,根据夜间灯光数据计算并修正各地级市内能源消费值,生成1km格网的河北省能源消费分布图.精度检验结果表明,能源消费的总体相对误差接近0,说明模型的模拟精度较高.全局空间自相关分析结果显示,2013年河北省能源消费具备显著的空间集聚性;局部空间自相关分析结果表明高能源消费区集中于各地城区和工业区,低能源消费区集中于以林地和耕地为主的北部和西部山区中,结果符合河北省能源消费的现状.     

Fault Diagnosis of Reciprocating Compressors Valve Based on Cyclostationary Method

The relationship between second-order cyclostationary method and time-frequency distribution is studied,and cyclic autocorrelation(CA) function is indicated to be one sort of special time-frequency distribution method.Furthermore,a fault diagnosis method for reciprocating compressors based on empirical mode decomposition (EMD) and CA function is proposed,and then it is applied to the fault diagnosis of reciprocating compressor valve.Firstly,the vibration signal of reciprocating compressor valve is decomposed by using EMD method,and several intrinsic mode functions (IMFs) are obtained.Secondly,the IMFs are evaluated by some denoising criterions to remove the noise and interfering ones.Finally,the CA functions of the remained IMFs are calculated,which will be used to reconstruct the CA function of the original vibration signal.Engineering application indicates that this method can sufficiently inhibit the cross-interference items of CA function.Therefore,more explicit working conditions of reciprocating compressor components can be achieved.     

How helpful are spatial effects in forecasting the growth of Chinese provinces?

In this paper, we make multi‐step forecasts of the annual growth rates of the real gross regional product (GRP) for each of the 31 Chinese provinces simultaneously. Beside the usual panel data models, we use panel models that explicitly account for spatial dependence between the GRP growth rates. In addition, the possibility of spatial effects being different for different groups of provinces (Interior and Coast) is allowed for. We find that both pooling and accounting for spatial effects help substantially to improve the forecast performance compared to the benchmark models estimated for each of the provinces separately. It is also shown that the effect of accounting for spatial dependence is even more pronounced at longer forecasting horizons (the forecast accuracy gain as measured by the root mean squared forecast error is about 8% at the 1‐year horizon and exceeds 25% at the 13‐ and 14‐year horizons). Copyright © 2010 John Wiley & Sons, Ltd.     

Spatial variability on vegetation parameters in Longitudinal Range-Gorge Region

Study on pattern and change on land cover stands an important aspect to global change. The spatial characteristics of ecological system show some familiar or disfamiliar principle with up-scaling or down-scaling process, which can be defined as spatial autocorrelation. In this paper, four typical parameters of ecological features, say Gross Primary Product （GPP）, Net Primary Product （NPP）, Normalized Difference Vegetation Index （NDVI）, and Enhanced Vegetation Index （EVI）, are selected to investigate the spatial variability on vegetative features in Longitudinal Range-Gorge Region （LRGR）, Southwest China. Moran coefficient （MC） is calculated for spatial autocorrelation degree, and semivariance function is used for spatial variability and spatial heterogeneity analysis. The results show that： （1） the ecological features show negative autocorrelation in the whole region. The higher anisotropy is on northsouth direction, and then on southeast-northwest direction. （2） The spatial variability scale is great on vegetative features in the whole region. （3） The autocorrelation distance on north-south direction is the greatest, then on southwest-northeast direction. Through the above analysis it can be found that the vegetative features are consistent with direction of landform in LRGR, which is shown on higher autocorrelation on south-north direction, and lower on west-east direction.     

中国股市收益率与波动性长记忆性的实证研究

根据 Granger关于序列长记忆性的定义 ,从股市收益率与波动性两个方面分析与研究了香港、上海和深圳股市的长记忆性 .结果表明 :中国股市收益率与波动性具有长记忆性 ,尽管收益率的长记忆性不如美国股市强 ;上海股市的记忆性明显强于深圳股市 ;分数可积模型较指数衰减自相关函数能更好地拟合样本的自相关系数