Modified Quasi‐Likelihood Ratio Test for Regime Switching

In this paper we propose a modified quasi‐likelihood ratio test of the null hypothesis of one regime against the alternative of two regimes in Markov regime‐switching models. The asymptotic distribution of the proposed test statistic is a simple function of Gaussian random variables, and the inference is no more complicated than in the standard case. Our simulations show that the proposed test has good finite sample size and power that are comparable to the quasi‐likelihood ratio test of Cho and White. We apply our test to stock returns and Japanese policy functions.     

The FMLS-based CUSUM statistic for testing the null of smooth time-varying cointegration in the presence of a structural break

In this paper we extend the FMLS-based CUSUM cointegration test (Xiao and Phillips, 2002) for testing the smooth time-varying cointegration null hypothesis. For this purpose we use Chebyshev time polynomials to specify time-varying coefficients under the null. We derive the limiting distribution of the statistic, which is pivotal with the order of the Chebyshev time polynomials, and we provide the critical values to conduct the proposed test.     

A CUSUM test for a long memory heterogeneous autoregressive model

A CUSUM test is proposed for testing structural breaks in a long-memory heterogeneous autoregressive model. The limiting distribution of the CUSUM test is shown to be a simple function of a standard Brownian bridge, contrasting with the nuisance parameter dependent asymptotics of other CUSUM tests based on fractional integration models. A Monte-Carlo experiment investigates finite sample size and power of the test. The proposed test is applied to a set of daily realized volatilities of the log-return of the Korean Won US Dollar exchange rate to reveal some evidence of a break in addition to a long-memory.     

Testing for Granger non-causality in heterogeneous panels

This paper proposes a very simple test of Granger (1969) non-causality for heterogeneous panel data models. Our test statistic is based on the individual Wald statistics of Granger non causality averaged across the cross-section units. First, this statistic is shown to converge sequentially to a standard normal distribution. Second, the semi-asymptotic distribution of the average statistic is characterized for a fixed T sample. A standardized statistic based on an approximation of the moments of Wald statistics is hence proposed. Third, Monte Carlo experiments show that our standardized panel statistics have very good small sample properties, even in the presence of cross-sectional dependence.     

EXAMINING THE ASYMMETRIC BEHAVIOUR OF MACROECONOMIC AGGREGATES IN ASIAN ECONOMIES

Abstract. The goal of this paper is to test for asymmetric behaviour of macroeconomic aggregates for three Asian economies; namely, Malaysia, Hong Kong and Korea. Whether macroeconomic aggregates can be characterised as asymmetric has important implications for policy‐making and econometric modelling including forecasting. We examine two forms of asymmetries; specifically deepness, which arises when a detrended time series contains an asymmetric distribution, and steepness, which arises when the first difference of a series contains an asymmetric distribution. Overall, our findings suggest that for all three countries, the bulk of the series display asymmetry behaviour.     

Modelling conditional heteroskedasticity: Application to the “IBEX-35” stock-return index

This paper compares alternative time-varying volatility models for daily stock-returns using data from Spanish equity index IBEX-35. Specifically, we estimate a parametric family of models of generalized autoregressive heteroskedasticity (which nests the most popular symmetric and asymmetric GARCH models), a semiparametric GARCH model, the generalized quadratic ARCH model, the stochastic volatility model, the Poisson Jump Diffusion model and, finally, a nonparametric model. Those models which use conditional standard deviation (specifically, TGARCH and AGARCH models) produce better fits than all other GARCH models. We also compare the within sample predictive power of all models using a standard efficiency test. Our results show that the asymmetric behaviour of responses is a statistically significant characteristic of these data. Moreover, we observe that specifications with a distribution which allows for fatter tails than a normal distribution do not necessarily outperform specifications with a normal distribution.     

Benford's Law and Fraud Detection: Facts and Legends

Abstract. Is Benford's law a good instrument to detect fraud in reports of statistical and scientific data? For a valid test, the probability of ‘false positives’ and ‘false negatives’ has to be low. However, it is very doubtful whether the Benford distribution is an appropriate tool to discriminate between manipulated and non‐manipulated estimates. Further research should focus more on the validity of the test and test results should be interpreted more carefully.     

Threshold effects in panel data stochastic frontier models of dairy production in Canada

One of the most enduring problems in econometrics is how to properly account for heterogeneity among firms. Threshold regression models are intuitively appealing methods to deal with this issue. We consider a fixed-effect panel data stochastic frontier model (Schmidt and Sickles, 1984; Martin-Marcos and Suarez-Galvez, 2000) and, relying on Hansen (1999, 2000a), we propose an estimator that accommodates multiple thresholds. Our model assumes absence of any unmeasured time invariant heterogeneity across firms as in Greene (2005, p. 277). Slope and threshold parameters can be estimated using a within estimator combined with a grid search over the threshold parameters. Testing for threshold effects is problematic because threshold parameters are not identified under the null hypothesis, a case of the so-called Davies' problem. We apply the bootstrap procedure proposed by Hansen (1999, 2000a) to test for the presence of thresholds. An asymptotic confidence set for the threshold parameter can be obtained by inverting an LR test, using the distribution result presented in Hansen (1999, 2000a). Our empirical application features a panel of Quebec dairy farms. We use farm size as the threshold variable. The presence of a trend in the specification matters for the determination of the number of thresholds. Technical efficiency scores and rankings of farms estimated from competing model specifications are highly correlated and do not vary significantly across groups of farm sizes defined by the threshold parameter values.     

Testing for bivariate stochastic dominance using inequality restrictions

We propose a test of bivariate stochastic dominance within a generalized framework for testing inequality restrictions, utilizing the covariance structure of the estimates of the joint distribution functions. Monte Carlo simulations and an empirical example assess its usefulness.     

On the distributional effects of income in an aggregate consumption relation

Abstract .  In this paper we analyse the influence of characteristics of the income distribution in modelling aggregate consumption expenditure. We model the aggregate consumption relation of a heterogeneous population, using a statistical distributional approach of aggregation, and apply it to UK-Family Expenditure Survey data. A bootstrap test based on a non-parametric estimation methodology, which accounts for the presence of continuous and discrete variables, suggests that the mean and the dispersion of the income distribution significantly influence aggregate consumption expenditure. Also, the parameters of the aggregate relation are time varying. These findings have implications for constructing empirically sound models of aggregate consumption expenditure.     

Market timing: Recent development and a new test

Henriksson-Merton's market timing test suffers nontrivial size distortion when the event forecast is autocorrelated. A new test is suggested to detect the dependence of two autocorrelated binary time series. It complements the existing tests due to better test power.     

Paradoxical price effects on insurance markets

In this paper we analyze a paradox phenomenon in certain insurance markets of some countries: in spite of there being strong competition, increasing prices can be found in some submarkets. As a reason for this, we need to note the special distribution of insurance products, a distribution primarily based on intermediaries — and, as a result, there is an increasing part for ‘intermediation’ costs within insurance products' cost structure due to competition for the intermediaries. We build an oligopolistic model assuming that the market is saturated, which is able to explain the overemployment of intermediaries and the high commissions involved. We analyze possible interventions and laws/regulations and suggest that the problem can be solved only with major regulation. Finally, we have come up with an empirical analysis to test the model.     

Some cautions on the use of nonlinear panel unit root tests: Evidence from a modified series-specific non-linear panel unit-root test

The purpose of this paper is to examine the relevance of applying nonlinear panel unit root test to examine the non-linear mean reversion behaviors of real exchange rates. We find that nonlinear panel unit root test may achieve lower power performance as compared to its alternative of linear panel unit test when the data generating process does not contain significant non-linear components. This finding post cautions to researchers in modeling and testing real exchanges behavior. We also develop a modified series-specific nonlinear panel unit root test and find evidence in favor of purchasing power parity hypothesis for China's four ASEAN trading partners in the period of February 1997 to August 2009.     

The Tobit cointegrated vector autoregressive model: An application to the currency market

The Tobit cointegrated vector autoregressive model proposed in this study extends the existing methodology by allowing the censored variable to be nonstationary. The approach requires deriving the distribution of the cointegration rank test and simulating new critical values. The empirical application refers to the currency market. It has confirmed that the exchange rate is driven by four main forces: inflation, terms of trade, the perception of the country-specific risk, and the state of the currency market. Temporary disequilibria in the currency market arise not only from the “fundamental” factors, but also from the contagion effect.     

Chaos in German stock returns — New evidence from the 0-1 test

This paper applies the 0-1 test for chaos to returns from the German stock market, providing empirical evidence of chaotic structures in the returns of all DAX members. For noise reduction purposes, wavelet denoising is employed prior to the application of the 0-1 test.     

On testing for nonlinearity in multivariate time series

This paper considers a multivariate extension of the test for neglected nonlinearity proposed by Tsay (1986) that uses principal components to overcome the problem of dimensionality that is common with tests of this type. Monte Carlo experiments reveal that the modified multivariate test provides a significant dimensional reduction without suffering from any systematic level distortion or power loss, and is more powerful than univariate nonlinearity tests.     

A nonparametric test for path dependence in discrete panel data

This paper proposes a test for path dependence in discrete panel data based on a characterization of stochastic processes that are mixtures of Markov chains. This test is applied to European Community Household Panel data on employment histories. The data allow to reject the null of no path dependence in all subsamples considered.     

A simple unit root test against asymmetric STAR nonlinearity with an application to real exchange rates in Nordic countries

Existing tests of the unit root hypothesis against the alternative hypothesis of exponential smooth transition autoregressive (ESTAR) nonlinearity implicitly assume symmetry under the alternative. This paper proposes a simple unit root test against the alternative of symmetric or asymmetric ESTAR nonlinearity. In the event that the unit root hypothesis is rejected, a simple test of symmetric versus asymmetric ESTAR nonlinearity is also proposed. The asymptotic distributions of the test statistics are straightforward to establish and finite-sample performance is studied with Monte Carlo simulations. An empirical application involving the real exchange rates of four Nordic countries against the U.S. dollar illustrates the usefulness of the new tests.     

A simple test for nonstationarity in mixed panels with incidental trends

Ng (2008) shows how the cross-sectional variance of the observed panel data can be used to construct a simple test for the proportion of non-stationary units. However, in the case with incidental trends the test is distorted. The present note shows how the distortions can be substantially reduced by the use of bias-adjustment. It also investigates the local power of the bias-adjusted test, which is shown to suffer from the same incidental trends problem previously only documented for conventional t$t$-tests.     

Hypothesis testing for arbitrary bounds

I derive a rigorous method to help determine whether a true parameter takes a value between two arbitrarily chosen points for a given level of confidence via a multiple testing procedure which strongly controls the familywise error rate. For any test size, the distance between the upper and lower bounds can be made smaller than that created by a confidence interval. The procedure is more powerful than other multiple testing methods that test the same hypothesis. This test can be used to provide an affirmative answer about the existence of a negligible effect.