Empiricial Comparison between Some Model Selection Criteria

Model selection aims to find the best model. Most of the usual criteria are based on goodness of fit and parsimony and aim to maximize a transformed version of likelihood. The situation is less clear when two models are equivalent: are they close to the unknown true model or are they far from it? Based on simulations, we study the results of Vuong's test, Cox's test, AIC and BIC and the ability of these four tests to discriminate between models.     

A new nonlinear asymmetric cointegration approach using error correction models

In this article, two new powerful tests for cointegration are proposed. The general idea is based on an intuitively appealing extension of the traditional, rather restrictive cointegration concept. In this article, we allow for a nonlinear, but most importantly a different, asymmetric convergence process to account for negative and positive changes in our cointegration approach. Using Monte Carlo simulations we verify, that the estimated size of the first test depends on the unknown value of a signal-to-noise ratio q. However, our second test—which is based on the original ideas of Kanioura and Turner—is more successful and robust in the sense that it works in all of the different evaluated situations. Furthermore it is shown to be more powerful than the traditional residual based Enders and Siklos method. The new optimal test is also applied in an empirical example in order to test for potential nonlinear asymmetric price transmission effects on the Swedish power market. We find that there is a higher propensity for power retailers to rapidly and systematically increase their retail electricity prices subsequent to increases in Nordpool's wholesale prices, than there is for them to reduce their prices subsequent to a drop in wholesale spot prices.     

Testing Wagner's Law in India: A cointegration and causality analysis

The present study empirically analyzes the validity of Wagner's Law for Indian economy. With the use of annual time series data from 1970–71 to 2013–14, all the six versions of Wagner's Law have been analyzed to test the relationship between government expenditure and gross domestic product. Wagner's Law states that the economic growth is the causative factor of the growth of the public expenditure. The study applied the unit root test and cointegration test to find the long-run relationship between government expenditure and gross domestic product. The present study employed the various econometric techniques such as unit root test, cointegration, and causality analysis for empirical analysis. The empirical analysis under study inferred mixed results of Wagner's Law for Indian economy, where four versions, namely Peacock, Gupta, Guffman, and Musgrave, found valid for Indian economy. The study concluded that the Wagner's Law is valid for the Indian economy except the Pryor and Mann Versions of the Wagner's Law.     

Inference after separated hypotheses testing: an empirical investigation for linear models

Model selection problems arise while constructing unbiased or asymptotically unbiased estimators of measures known as discrepancies to find the best model. Most of the usual criteria are based on goodness-of-fit and parsimony. They aim to maximize a transformed version of likelihood. For linear regression models with normally distributed error, the situation is less clear when two models are equivalent: are they close to or far from the unknown true model? In this work, based on stochastic simulation and parametric simulation, we study the results of Vuong's test, Cox's test, Akaike's information criterion, Bayesian information criterion, Kullback information criterion and bias corrected Kullback information criterion and the ability of these tests to discriminate between non-nested linear models.     

Vector autoregression and causality: a theoretical overview and simulation study

This paper provides a theoretical overview of Wald tests for Granger causality in levels vector autoregressions (VAR's) and Johansen-type error correction models (ECM's). The theory is based on results in Toda and Phillips (1991a) and allows for stochastic and deterministic trends as well as arbitrary degrees of cointegration. We recommend some operational procedures for conducting Granger causality tests that are based on the Gaussian maximum likelihood estimation of ECM's. These procedures are applicable in the important practical case of testing the causal effects of one variable on another group of variables and vice versa. This paper also investigates the sampling properties of these testing procedures through simulation exercises. Three sequential causality tests in ECM's are compared with conventional causality tests in levels and differences VAR's.     

Granger因果关系检验的模型选择与相关策略研究

运用Granger因果关系检验识别确定经济变量间因果关系是经济研究中极为常见的分析模式,然而在具体应用时,Granger因果关系检验的功效会受到模型形式选择与检验策略因素的影响,为此,解析了Granger因果关系检验的水平型VAR、差分型VAR、VEC三种模型形式选择的基本原理,探讨了与模型选择相关的四大检验策略,即变量个数选择、滞后阶数选择、变量单整性检验、协整空间维数选择,并给出了Granger因果关系检验相对稳妥的实践操作程序。     

No-cointegration test based on fractional differencing: Some Monte Carlo results

This paper examines the use of the t-statistic in the Geweke–Porter-Hudak regression for the estimation of the fractional differencing parameter as a test for cointegration. The critical values of the test statistic are estimated using Monte Carlo methods. The results confirm that the test will over-reject the null hypothesis of no cointegration if the standard-normal critical values are used. The estimated critical values are generally robust to the nuisance parameters in the autoregressive or moving average specification of the error process of the component time series. Exceptions occur when the dependent variable in the cointegration regression follows an autoregressive process with a large positive parameter or a moving average process with a large negative parameter.     

Testing for cointegration at any frequency using spectral methods

Summary In this paper a new simple test for cointegration at any frequency is presented. This method can thus be applied to test for cointegration both at the zero and at the seasonal frequencies. It requires the estimation of the coherency spectrum of weakly stationary processes, therefore only standard spectral theory is involved. The testing procedure is similar to the one suggested by Phillips and Ouliaris (1988) and recently generalized by Joyeux (1992) to frequencies different from zero, but it does not suffer of some problems connected with the use of principal components methods in the frequency domain. Invited paper at the Conference held in Bologna, Italy, 27–28 May 1993, on ?Statistical Tests: Methodology and Econometric Applications?.     

非线性协整检验与“费雪效应”机制分析

本文分别在线性Engle-Granger协整模型和非线性指数平滑迁移自回归误差修正模型 (ESTAR-ECM) 的框架下,对我国名义利率与通货膨胀率序列进行了长期均衡关系的检验。发现线性协整模型不能捕捉到我国名义利率与通货膨胀率的长期均衡关系,而对于ESTAR-ECM模型,无论利用商业银行1年期贷款利率还是7天期银行间同业拆借利率作为名义利率的代理变量,均证实名义利率与通货膨胀率具有长期稳定的均衡关系,表明“费雪效应”在我国是成立的。但由于“费雪效应”系数小于1,表明名义利率与通货膨胀率之间仅存在弱的“费雪效应”。其意义在于,我国利率政策对稳定通胀预期、抑制通货膨胀具有一定的正面效应,但由于利率对通货膨胀反应不足,导致完全依靠利率政策控制目前较高的通货膨胀有一定的困难。     

Bayesian Inference in Cointegrated I (2) Systems: A Generalization of the Triangular Model

This paper generalizes the cointegrating model of Phillips (1991) to allow for I (0), I (1) and I (2) processes. The model has a simple form that permits a wider range of I (2) processes than are usually considered, including a more flexible form of polynomial cointegration. Further, the specification relaxes restrictions identified by Phillips (1991) on the I (1) and I (2) cointegrating vectors and restrictions on how the stochastic trends enter the system. To date there has been little work on Bayesian I (2) analysis and so this paper attempts to address this gap in the literature. A method of Bayesian inference in potentially I (2) processes is presented with application to Australian money demand using a Jeffreys prior and a shrinkage prior.     

中国费雪效应的门限协整检验

由于中国费雪效应的研究结果具有很大的不一致性,结合中国1991年1月至2008年12月之间的数据,应用可以刻画变量间非线性均衡关系的门限协整理论检验费雪效应,研究结果显示：第一,中国的名义利率与通货膨胀率均为单位根过程,二者之间不存在线性协整关系,而是存在两个门限值的门限协整关系;第二,当通货膨胀率小于-0．8％时,中国费雪效应不存在,而当通货膨胀率在-0．8％～12．03％2;间时,中国存在值为0．42的部分费雪效应;当通货膨胀率大于12．03％时,中国存在值为0．05的部分费雪效应。     

Tests for Parameter Instability in Regressions with 1(1) Processes

This article derives the large-sample distributions of Lagrange multiplier (LM) tests for parameter instability against several alternatives of interest in the context of cointegrated regression models. The fully modified estimator of Phillips and Hansen is extended to cover general models with stochastic and deterministic trends. The test statistics considered include the SupF test of Quandt, as well as the LM tests of Nyblom and of Nabeya and Tanaka. It is found that the asymptotic distributions depend on the nature of the regressor processes—that is, if the regressors are stochastic or deterministic trends. The distributions are noticeably different from the distributions when the data are weakly dependent. It is also found that the lack of cointegration is a special case of the alternative hypothesis considered (an unstable intercept), so the tests proposed here may also be viewed as a test of the null of cointegration against the alternative of no cointegration. The tests are applied to three data sets—an aggregate consumption function, a present value model of stock prices and dividends, and the term structure of interest rates.     

我国费雪效应的非参数检验

本文基于我国1990:01—2007:04期间的名义利率与通货膨胀率月度数据非线性变化的特征,应用非参数单位根和非参数协整理论检验我国是否存在费雪效应, 进而应用非参数局部线性变窗宽估计计算我国的费雪系数。由此产生的结论为：第一,非参数单位根检验发现我国名义利率与通货膨胀率都是非平稳的单位根过程;第二,非参数协整检验的结论为, 我国名义利率与通胀变化率之间存在长期的非线性协整关系, 这一结论表明我国至少存在弱的费雪效应;第三,非参数局部线性变窗宽估计计算的费雪效应(系数)的均值为0.4055,这一结果进一步支持我国存在弱的费雪效应,其隐含的意义为,当前加息对稳定通胀将产生正面效应,进一步, 如适时适度的调整利率, 很可能抑制当前较高的CPI向高通胀的转化。     

The effects of additive outliers on the seasonal KPSS test: a Monte Carlo analysis

This article builds on the test proposed by Lyhagen [The seasonal KPSS statistic, Econom. Bull. 3 (2006), pp. 1–9] for seasonal time series and having the null hypothesis of level stationarity against the alternative of unit root behaviour at some or all of the zero and seasonal frequencies. This new test is qualified as seasonal-frequency Kwiatkowski–Phillips–Schmidt–Shin (KPSS) test and it is not originally supported by a regression framework.

The purpose of this paper is twofold. Firstly, we propose a model-based regression method and provide a clear illustration of Lyhagen's test and we establish its asymptotic theory in the time domain. Secondly, we use the Monte Carlo method to study the finite-sample performance of the seasonal KPSS test in the presence of additive outliers. Our simulation analysis shows that this test is robust to the magnitude and the number of outliers and the statistical results obtained cast an overall good performance of the test finite-sample properties.     

An Adaptive Method of Variable Selection in Regression

An adaptive variable selection procedure is proposed which uses an adaptive test along with a stepwise procedure to select variables for a multiple regression model. We compared this adaptive stepwise procedure to methods that use Akaike's information criterion, Schwartz's information criterion, and Sawa's information criterion. The simulation studies demonstrated that the adaptive stepwise method is more effective than the traditional variable selection methods if the error distribution is not normally distributed. If the error distribution is known to be normally distributed, the variable selection method based on Sawa's information criteria appears to be superior to the other methods. Unless the error distribution is known to be normally distributed, the adaptive stepwise method is recommended.     

A bivariate fractionally cointegrated relationship in the context of cyclical structures

This paper deals with the analysis of cointegration in a bivariate system. However, we depart from the classic concept of cointegration in two aspects. First, we permit fractional degrees of integration in both the parent series and in their linear combination. Second, instead of assuming that the pole or singularity in the spectrum takes places at the zero frequency, we consider the case where the singularity occurs at a frequency λ in the interval (0, π]. We use a procedure that follows the same lines as the two-step testing strategy of R.F. Engle, and C.W.J. Granger, [Cointegration and error correction model. Representation, estimation and testing, Econometrica 55 (1987), pp. 251–276]. Thus, we test first the order of integration in the individual series, which are specified in terms of the Gegenbauer polynomials. Then, if the two series share the same degree of integration at a given frequency, we test the null hypothesis of no cointegration against the alternative of fractional cyclical cointegration, by testing the order of integration on the estimated residuals from the cointegrating regression. Finite sample critical values are obtained, and the power properties of the test are examined. An empirical application is also carried out at the end of the article.     

Hypotheses testing for generalized order statistics with simple order restrictions on model parameters under the alternative

In applications of generalized order statistics as, for instance, reliability analysis of engineering systems, prior knowledge about the order of the underlying model parameters is often available and may therefore be incorporated in inferential procedures. Taking this information into account, we establish the likelihood ratio test, Rao's score test, and Wald's test for test problems arising from the question of appropriate model selection for ordered data, where simple order restrictions are put on the parameters under the alternative hypothesis. For simple and composite null hypothesis, explicit representations of the corresponding test statistics are obtained along with some properties and their asymptotic distributions. A simulation study is carried out to compare the order restricted tests in terms of their power. In the set-up considered, the adapted tests significantly improve the power of the associated omnibus versions for small sample sizes, especially when testing a composite null hypothesis.     

MULTIPLE COMPARISONS OF LOG-LIKELIHOODS AND COMBINING NONNESTED MODELS WITH APPLICATIONS TO PHYLOGENETIC TREE SELECTION

We consider multiple comparisons of log-likelihood's to take account of the multiplicity of testings in selection of nonnested models. A resampling version of the Gupta procedure for the selection problem is used to obtain a set of good models, which are not significantly worse than the maximum likelihood model; i.e., a confidence set of models. Our method is to test which model is better than the other, while the object of the classical testing methods is to find the correct model. Thus the null hypotheses behind these two approaches are very different. Our method and the other commonly used approaches, such as the approximate Bayesian posterior, the bootstrap selection probability, and the LR test against the full model, are applied to the selection of molecular phylogenetic tree of mammal species. Tree selection is a version of the model-based clustering, which is an example of nonnested model selection. It is shown that the structure of the tree selection problem is equivalent to that of the variable selection problem of the multiple regression with some constraints on the combinations of the variables. It turns out that the LR test rejects all the possible trees because of the misspecification of the models, whereas our method gives a reasonable confidence set. For a better understanding of the uncertainty in the selection, we combine the maximum likelihood estimates (MLE's) of the trees to obtain the full model that includes the trees as the submodels by using a linear approximation of the parametric models. The MLE of the phylogeny is then represented as a network of species rather than a tree. A geometrical interpretation of the problem is also discussed.     

Model selection criteria for reduced rank multivariate time series: a simulation study

The main focus of our paper is to compare the performance of different model selection criteria used for multivariate reduced rank time series. We consider one of the most commonly used reduced rank model, that is, the reduced rank vector autoregression (RRVAR (p, r)) introduced by Velu et al. [Reduced rank models for multiple time series. Biometrika. 1986;7(31):105–118]. In our study, the most popular model selection criteria are included. The criteria are divided into two groups, that is, simultaneous selection and two-step selection criteria, accordingly. Methods from the former group select both an autoregressive order p and a rank r simultaneously, while in the case of two-step criteria, first an optimal order p is chosen (using model selection criteria intended for the unrestricted VAR model) and then an optimal rank r of coefficient matrices is selected (e.g. by means of sequential testing). Considered model selection criteria include well-known information criteria (such as Akaike information criterion, Schwarz criterion, Hannan–Quinn criterion, etc.) as well as widely used sequential tests (e.g. the Bartlett test) and the bootstrap method. An extensive simulation study is carried out in order to investigate the efficiency of all model selection criteria included in our study. The analysis takes into account 34 methods, including 6 simultaneous methods and 28 two-step approaches, accordingly. In order to carefully analyse how different factors affect performance of model selection criteria, we consider over 150 simulation settings. In particular, we investigate the influence of the following factors: time series dimension, different covariance structure, different level of correlation among components and different level of noise (variance). Moreover, we analyse the prediction accuracy concerned with the application of the RRVAR model and compare it with results obtained for the unrestricted vector autoregression. In this paper, we also present a real data application of model selection criteria for the RRVAR model using the Polish macroeconomic time series data observed in the period 1997–2007.     

Threshold Cointegration and the PPP Hypothesis

Self-Exciting Threshold Autoregressive (SETAR) models are a non-linear variant of conventional linear Autoregressive (AR) models. One advantage of SETAR models over conventional AR models lies in its flexible nature in dealing with possible asymmetric behaviour of economic variables. The concept of threshold cointegration implies that the Error Correction Mechanism (ECM) at a particular interval is inactive as a result of adjustment costs, and active when deviations from equilibrium exceed certain thresholds. For instance, the presence of adjustment costs can, in many circumstances, justify the fact that economic agents intervene to recalibrate back to a tolerable limit, as in the case when the benefits of adjustment are superior to its costs. We introduce an approach that accounts for potential asymmetry and we investigate the presence of the relative version of the purchasing power parity (PPP) hypothesis for 14 countries. Based on a threshold cointegration adaptation of the unit root test procedure suggested by Caner & Hansen (2001), we find evidence of an asymmetric adjustment for the relative version of PPP for eight pairs of countries.