基于结构突变和截面相关的面板协整检验

本文基于Westerlund和Edgerton（2008）,考虑了无时间趋势和有时间趋势的面板协整检验。在检验协整时,本文不仅允许误差项存在异方差、序列相关以及截面相关,而且还允许各截面在截距和协整斜率上存在未知时点的多个突变点。蒙特卡洛模拟结果表明,（1）该检验的具有较小的水平扭曲和较高的检验势,（2）将模型拓展到不含有趋势项的情形是必要的。在此基础上,使用基于动态最小二乘估计量的新统计量对国际CO2排放和经济增长关系进行检验,发现在考虑了突变和截面相关的情形下,两者间的长期均衡关系确实存在。     

具有GARCH(1,1)-Normal-errors的单位根过程DF检验的可靠性研究

本文指出了Kim和Schmidt（1993）等在研究GARCH-errors对DF单位根检验有限样本性质影响时存在的方法上的缺陷。通过理论分析和Monte Carlo随机模拟，发现对于具有GARCH（1，1）-Normal-errors的单位根过程采用DF统计量进行检验应遵循以下规律：（1）对于任意给定的初始条件方差h0和条件方差方程的常数项ω，当去掉初始生成的数据足够多时，可以得到相当平稳的误差项序列，并且h0和ω对DF统计量k和r分布的影响可以忽略不计；（2）无论采用DF检验的临界值还是采用统计量的实际分位数，k检验均比τ检验具有较高的可靠性；（3）对于给定的干扰项系数，条件方差方程的系数和越高，k检验和τ检验的可靠性越差。     

Enders-Granger方法在协整检验中的应用研究

本文将协整检验由传统的线性协整检验扩展到线性协整检验和阈值协整检验,并在Enders和Granger(1998)方法的基础上提出了一个新的检验协整是否存在的Sup-F和Sup-F*统计量。通过MC仿真研究发现:在线性协整下,ADF方法比Sup-F法具有更高的检验势,但在"持久性"较强时,Sup-F检验比ADF检验法具有更高的检验势;Sup-F统计量在Three-Regime的阈值协整检验中比ADF法有更高的检验势;Sup-F*在检验协整(包括线性协整和阈值协整)时都具有较低的检验势;随着在不同Regime中自回归系数差距的增大(非对称程度增大),sup-F统计量的检验势提高很快,且比ADF法的检验势高。     

同期相关面板数据退势单位根检验的小样本性质

本文基于SUR回归将时间序列的两种单位根检验（ADF—GLS检验）推广到面板数据，得到了同期相关面板数据退势单位根检验，称为SUR—ADF—GLS检验。通过蒙特卡洛试验研究发现，SUR—ADF—GLS检验具有良好的小样本性质。并且，SUR—ADF—GLS检验关于面板数据的同期相关性结构存在着较强的“依存性”。     

修正的倒向上确界ADF泡沫检验方法——来自上证综指的证据

模拟蒙特卡洛分析发现,当两个泡沫间隔很短或第一个泡沫比第二个持续时间较长时,BSADF检验方法不能准确地估计出第二个泡沫,而本文修正的BSADF检验方法能有效地克服此问题。通过比较这两种方法对上证综指泡沫的产生和破灭时间的估计差异,发现BSADF检验只能检验出一个泡沫,泡沫持续期为2006年12月至2007年12月,而修正的BSADF发现了紧邻的第二个短暂性泡沫,发生在金融危机期间。     

广义Breitung非参数单位根检验研究——基于平稳假设下序列长期方差一致估计的分析

Breitung检验中生成序列的误差项的自相关会影响有限样本性质。本文用平稳假设下序列长期方差的一致估计量作为统计量的分母对其进行了修正。给出了修正后的统计量及其渐近理论,并对修正前后的有限样本性质进行了仿真。结果显示,修正后统计量概率密度的左偏有所减少;当误差项有自相关时,修正后检验的水平扭曲有所改进;当样本较小时,随误差项自回归(移动平均)系数或序列自回归系数的增加,修正后检验的势逐渐大于Breitung检验的势。     

固定效应模型空间相关性的Bootstrap LM-error检验

基于Lee和Yu(2010)的正交转换消除固定效应,将FDB方法用于空间固定效应模型误差自相关的LM-error检验。在不同的误差结构、样本量、空间权重矩阵、序列相关系数和固定效应大小条件下,比较渐近LM-error检验和Bootstrap LM-error检验的水平扭曲和功效。蒙特卡洛模拟实验表明,当误差项为标准正态分布时,两者检验均具有较好的水平扭曲和功效表现。当误差项为异方差或者序列相关时,渐近LM-error检验存在严重的水平扭曲,而Bootstrap LM-error检验能够有效地校正其水平扭曲,且其检验功效与渐近LM-error检验功效近似相等,Bootstrap LM-error检验是更为理想的检验方法。     

ADF单位根检验法的替代方法

研究目标：ADF单位根检验法的替代方法——消除趋势检验法。研究方法：通过蒙特卡罗仿真估算了消除趋势法、ADF法和差分法的检验临界值和检验功效;估算了三种方法检验临界值关于样本容量的响应面函数;估算了消除趋势回归中趋势项和截距项的t统计量的检验临界值及其响应面函数。研究发现：在各种情形下,消除趋势法的检验功效均大于ADF法,而且在误差项相关时更为明显,说明消除趋势法优于ADF法,可以完全替代ADF法;在某些情况下,差分法的检验功效较大。研究创新：对ADF检验模型进行了改进,提出了消除趋势检验法。研究价值：在单位根检验时,可以用消除趋势法代替一直以来被广泛采用的ADF法,提高单位根检验的功效。     

非参数统计视角下的异方差检验设计及其应用

以数据生成过程为导向,探讨了异方差来源的基本类型;依据非参数统计的基本思想,设计了切实可行的Mood方差检验方法与平方秩检验方法,并针对异方差来源类型,分析了相应的检验思路。蒙特卡罗模拟表明,Mood方差检验方法在异方差检验方面具有很高的检验效力;平方秩检验方法在异常经济现象情形下检验效力较低,而在其他情形下检验效力很高。同时,进一步阐释了纠正异方差的基本逻辑。     

泡沫分析视角下的指数型基金抗风险能力研究

立足泡沫分析视角,研究了金融产品的抗风险能力,选择了能代表某类金融市场的指数型基金,引入了金融物理学中的LPPL模型,进而分别从理论和实证两方面对指数型基金的抗风险能力进行了分析,对金融产品抗风险能力以及金融市场泡沫的相关性与内涵性进行了解释。研究结论表明：我国金融市场的抗风险能力相对较弱,无论是市场指数还是指数基金均存在新一轮泡沫破灭的可能;相比市场指数,指数型基金呈现更加多样的泡沫形式,其抗风险能力相对较强;从泡沫破灭缓冲期对比来看,我国指数型基金抗风险能力优于市场指数。     

Detecting periodically collapsing bubbles: a Markov-switching unit root test

This paper addresses the problem of testing for the presence of a stochastic bubble in a time series in the case that the bubble is periodically collapsing so that the asset price keeps returning to the level implied by the market fundamentals. As this is essentially a problem of identifying the collapsing periods from the expanding ones, we propose using a generalization of the Dickey–Fuller test procedure which makes use of the class of Markov regime-switching models. The potential of the new methodology is illustrated via simulation, and an empirical example is given. Copyright © 1999 John Wiley & Sons, Ltd.     

Bubbles detection for inter-war European hyperinflation: A threshold cointegration approach

This paper undertakes an empirical investigation into the existence of inflationary bubbles during the inter-war European hyperinflation for Germany, Hungary, Poland and Russia. Our Monte Carlo simulations show that the residual-based threshold cointegration methodology of Caner and Hansen (2001) is better able to detect periodically collapsing bubbles. Moreover, this methodology possesses greater power against nonlinear stationary alternatives in a finite sample than several commonly used cointegration tests that do not allow for multiple regime shifts. The empirical results of the threshold cointegration tests provide evidence of stationary, regimeswitching processes in money demand dynamics, but suggest that there are no inflationary bubbles in any of the countries.     

Diagnosing housing fever with an econometric thermometer

Housing fever is a popular term to describe an overheated housing market or housing price bubble. Like other financial asset bubbles, housing fever can inflict harm on the real economy, as indeed the U.S. housing bubble did in the period following 2006 leading up to the general financial crisis and great recession. One contribution that econometricians can make to minimize the harm created by a housing bubble is to provide a quantitative “thermometer” for diagnosing ongoing housing fever. Early diagnosis can enable prompt and effective policy action that reduces long-term damage to the real economy. This paper provides a selective review of the relevant literature on econometric methods for identifying housing bubbles together with some new methods of research and an empirical application. We first present a technical definition of a housing bubble that facilitates empirical work and discuss significant difficulties encountered in practical work and the solutions that have been proposed in the past literature. A major challenge in all econometric identification procedures is to assess prices in relation to fundamentals, which requires measurement of fundamentals. One solution to address this challenge is to estimate the fundamental component from an underlying structural relationship involving measurable variables. A second aim of the paper is to improve the estimation accuracy of fundamentals by means of an easy-to-implement reduced-form approach. Since many of the relevant variables that determine fundamentals are nonstationary and interdependent we use the endogenous instrumental variable based method (IVX) to estimate the reduced-form model to reduce the finite sample bias which arises from highly persistent regressors and endogeneity. The recursive evolving test proposed by Phillips, Shi, and Yu (PSY) is applied to the estimated nonfundamental component for the identification of speculative bubbles. The new bubble test developed here is referred to as PSY-IVX. An empirical application to the eight Australian capital city housing markets over the period 1999–2017 shows that bubble testing results are sensitive to different ways of controlling for fundamentals and highlights the importance of accurate estimation of these housing market fundamentals.     

中国十大宜居城市房地产价格泡沫度的实证研究

本文通过单位根ADF检验,对我国十大宜居城市中的九个城市（剔除桂林,数据不可得）的房地产价格泡沫进行了初步检验,结果显示除青岛外其他城市房地产价格都面临着不同程度的泡沫。然后通过建立时变风险溢价模型和三变量VAR模型,对除青岛以外的剩余城市的房地产价格泡沫进行了度量,结果显示：上海、大连、北京、广州、成都、杭州、珠海和厦门的房地产市场都存在着不同程度的泡沫,其中北京、杭州和珠海的房地产市场泡沫已经达到了30％以上,而上海、大连和厦门的房地产市场泡沫在20％以上,成都和广州的房地产市场泡沫也在10％以上。使用该方法可以度量任何一个地区、省份、市区的房地产价格泡沫度,可以为房地产商、购房人和中央及地方政府提供相应的决策依据。     

中国房地产市场价格泡沫的检验与成因机理研究

本文使用Pesaran(2007)提出的CIPS面板单位根检验和Pedroni面板协整检验,对我国房地产价格是否存在泡沫等问题进行了研究。检验结果表明,房屋销售价格指数和租赁价格指数序列的单整阶数不相等,并且二者不存在协整关系,因此在样本期间内我国房地产价格存在泡沫。在此基础上,本文对我国房地产价格泡沫产生的原因进行了简要分析,并给出相应的政策建议。     

Two-stage non Gaussian QML estimation of GARCH models and testing the efficiency of the Gaussian QMLE

In generalized autoregressive conditional heteroskedastic (GARCH) models, the standard identifiability assumption that the variance of the iid process is equal to 1 can be replaced by an alternative moment assumption. We show that, for estimating the original specification based on the standard identifiability assumption, efficiency gains can be expected from using a quasi-maximum likelihood (QML) estimator based on a non Gaussian density and a reparameterization based on an alternative identifiability assumption. A test allowing to determine whether a reparameterization is needed, that is, whether the more efficient QMLE is obtained with a non Gaussian density, is proposed.     

Systemic risk of Chinese financial institutions and asset price bubbles

Although there has not been a large-scale systemic crisis in China, high-risk financial events have occurred continuously in recent years. This research thus creatively analyzes the determinants of systemic risk for Chinese financial institutions from the view of asset price bubbles. First, we identify bubbles in the China stock and real estate markets on the basis of the generalized sup Augmented Dickey-Fuller (GSADF) model and explain the reasons for bubble formations according to the stage of China's economic development and policies implementation. At this stage, considering the differences in economic development levels of different cities, the real estate bubbles in the first, second and third tier cities and the whole country were innovatively identified. Second, on the basis of the DCG-GARCH-CoVaR model to measure the systemic risk of listed financial institutions in China and to classify institutions, the results show that the main source of such risk is the banking sector. Furthermore, by constructing regression models, stock market bubbles and real estate bubbles both positively correlate with systemic risk throughout the sample period. Meanwhile, the impact of bubbles on the systemic risk of different types of financial institutions was taken into account so that regulators prioritized different types of institutions with different characteristics when faced with decisions. Finally, we provide macro-prudential policy advice to regulators in order to weaken the impact of bubbles on financial stability to avoid systemic crises.     

齐次及非齐次马氏域变模型在股价泡沫检验中的应用

本文建立非齐次马氏域变模型检验股市价格泡沫,并结合对1996年1月至2010年6月间我国股市价格泡沫程度的实证度量,与齐次马氏域变模型的检验效果进行比较。两模型都能在一定程度上反映我国股票市场价格在各时间段的不同泡沫水平,但非齐次马氏域变模型所得到有泡沫概率和无泡沫概率区分度更高,比齐次马氏域变模型能更加精确的区分各时点上的泡沫水平。实证结果也表明,近期我国应对金融危机的超宽松经济政策对资本市场的刺激作用较为直接,而对实体经济的作用相对滞后,股市泡沫水平再度上升。     

Multi-country tests for price level bubbles

Flood and Garber (1980) have examined data from the German hyperinflation to test for a price level bubble. In this paper, we extend this research to other episodes of hyperinflation. By testing for a simultaneous bubble across countries, we implement the conceptual experiment proposed by Flood and Garber to derive asymptotic distributions for the test statistics.     

A re-examination of housing bubbles: Evidence from European countries

This study uses data from six Eurozone countries and the United Kingdom between 1980Q1 and 2018Q4 to examine whether these countries had housing bubbles during the observed period. Whereas typical studies make strictly limited assumptions regarding interest rates, we make an unconventional argument for the necessity of testing the integration relationship between the price–rent ratio and the interest rate reciprocal to determine the existence of housing bubbles. To verify this study’s proposition, two housing bubble indicators were adopted to dynamically examine periods of housing bubbles in European countries by using a series of individual countries and panel data from Eurozone countries. According to the empirical results for individual countries, although the price–rent ratio indicates the occurrence of housing booms in the targeted countries, the evidence for housing bubbles is unclear. The dynamic bubble indicator revealed that housing bubbles occurred in France and Ireland within a short period in 1993Q3 and 2000Q2, respectively. Spain experienced two short-term housing bubbles in 1990Q1 and 2015Q1. The short-term bubbles signify that the housing markets were efficient. Once the price–rent ratio failed to converge toward the nominal interest rate, market traders’ rational behavior can immediately correct the short-term market divergence. The panel data of the Eurozone countries also reveals that simply using the price–rent ratio for examination may underestimate the correction of the housing markets. In conclusion, the results of this study demonstrate the importance of the interest rate in controlling the housing market.