A Monte Carlo Analysis of the Fisher Randomization Technique: Reviving Randomization for Experimental Economists

Data created in a controlled laboratory setting are a relatively new phenomenon to economists. Traditional data analysis methods using either parametric or nonparametric tests are not necessarily the best option available to economists analyzing laboratory data. In 1935, Fisher proposed the randomization technique as an alternative data analysis method when examining treatment effects. The observed data are used to create a test statistic. Then treatment labels are shuffled across the data and the test statistic is recalculated. The original statistic can be ranked against all possible test statistics that can be generated by these data, and a p-value can be obtained. A Monte Carlo analysis of t-test, the Mann-Whitney U-test, and the exact randomization t-test is conducted. The exact randomization t-test compares favorably to the other two tests both in terms of size and power. Given the limited distributional assumptions necessary for implementation of the exact randomization test, these results suggest that experimental economists should consider using the exact randomization test more often. This revised version was published online in August 2006 with corrections to the Cover Date.     

Panel unit root tests under cross section dependence with recursive mean adjustment

Utilizing recursive mean adjustment (RMA) we provide two unit root tests: the covariate RMA unit root test and the panel feasible generalized RMA unit root test. The proposed panel unit root tests are precise and powerful, especially when T > N.     

Can individuals’ beliefs help us understand nonadherence to malaria test results? Evidence from rural Kenya

In malaria‐endemic countries about a quarter of test‐negative individuals take antimalarials (artemisinin‐based combination therapies [ACTs]). ACT overuse depletes scarce resources for subsidies and contributes to parasite resistance. As part of an experiment in Kenya that provided subsidies for rapid diagnostic test and/or for ACTs conditionally on being positive, we studied the association between beliefs on malaria status (prior and posterior the intervention) and decisions to get tested and to purchase ACTs. We find that prior beliefs do not explain the decision of getting tested (conditional on the price) and nonadherence to a negative test. However, test‐negative individuals who purchase ACTs report higher posterior beliefs than those who do not, consistent with a framework in which the formers revise beliefs upward, while the latters do not change or revise downward. We also do not find evidence that prior beliefs on ACT effectiveness and trust in test results play any major role in explaining testing or treatment behavior. Further research is needed to improve adherence to malaria‐negative test results.     

A new unit root test based on F-statistic in ESTAR framework

This article proposes a new F-type unit test in the exponential smooth transition autoregressive framework. We derive the asymptotic nonstandard distribution of the proposed test and explore its finite sample properties; simulation results show our test has greater power than the t_kss test proposed by Kapetanios et al.(2003). Finally, an application on the real exchange rates further underpins its superiority.     

IMMIGRATION UNEMPLOYMENT RELATIONSHIP: THE EVIDENCE FROM CANADA*

This paper examines the relationship between unemployment and immigration in Canada. The bi‐directional causality test finds no evidence of a significant effect of Canadian immigration on unemployment. Cointegration tests indicate that there is no observed increase in aggregate unemployment due to immigration in the long run. The results from the causality test based on the vector error correction model confirm that, in the short run, past unemployment does cause (less) immigration but not vice versa. There is also a long‐run positive relationship among per‐capita GDP, immigration rate and real wages. The results indicate that, in the short‐run, more immigration is possibly associated with attractive Canadian immigration policies, and in the long‐run, as the labour market adjusts, Canadian‐born workers are likely to benefit from increased migration.     

Bootstrapping the autoregressive distributed lag test for cointegration

We propose a bootstrap autoregressive-distributed lag (ARDL) test. By applying the appropriate bootstrap method, some weaknesses underlying the Pesaran, Shin and Smith ARDL bounds test are addressed including size and power properties and the elimination of inconclusive inferences. In addition, inferences based solely on the significance of the F-test and single t-test from the ARDL bounds test are not sufficient to avoid degenerate cases. The bootstrap ARDL test provides an additional test on the significance of coefficients on lagged levels of the regressors, which provides a better insight into the cointegration status of the model.     

TESTING FOR STATIONARITY IN HETEROGENEOUS PANEL DATA IN THE CASE OF MODEL MISSPECIFICATION

This paper investigates the performance of the tests proposed by Hadri and by Hadri and Larsson for testing for stationarity in heterogeneous panel data under model misspecification. The panel tests are based on the well known KPSS test (cf. Kwiatkowski et al.) which considers two models: stationarity around a deterministic level and stationarity around a deterministic trend. There is no study, as far as we know, on the statistical properties of the test when the wrong model is used. We also consider the case of the simultaneous presence of the two types of models in a panel. We employ two asymptotics: joint asymptotic, T, N →∞ simultaneously, and T fixed and N allowed to grow indefinitely. We use Monte Carlo experiments to investigate the effects of misspecification in sample sizes usually used in practice. The results indicate that the assumption that T is fixed rather than asymptotic leads to tests that have less size distortions, particularly for relatively small T with large N panels (micro‐panels) than the tests derived under the joint asymptotics. We also find that choosing a deterministic trend when a deterministic level is true does not significantly affect the properties of the test. But, choosing a deterministic level when a deterministic trend is true leads to extreme over‐rejections. Therefore, when unsure about which model has generated the data, it is suggested to use the model with a trend. We also propose a new statistic for testing for stationarity in mixed panel data where the mixture is known. The performance of this new test is very good for both cases of T asymptotic and T fixed. The statistic for T asymptotic is slightly undersized when T is very small (≤10).     

ASIAN REAL EXCHANGE RATES AND OIL PRICES: A COINTEGRATION ANALYSIS UNDER STRUCTURAL BREAKS

We examine the long‐run relationship between Asian real exchange rates and oil prices in the presence of structural breaks. The relevance of considering breaks is demonstrated by utilizing the Johansen et al. procedure that allows for up to two predetermined breaks. Using conventional tests that do not consider breaks reveals no evidence of cointegration. However, the Johansen et al. procedure clearly demonstrates the importance of considering breaks and provides strong support for a stable long‐run relationship in all but Japan and the Philippines. Moreover, the results suggest evidence of bi‐directional causality in Malaysia and Thailand, uni‐directional causality from exchange rates to oil prices in Korea, the Philippines, and Singapore, uni‐directional causality from oil prices to the exchange rate in Indonesia, and no evidence of causality in Japan.     

Identifying cost-effective screening algorithms for active hepatitis C virus infections in a high prevalence setting

Objective: To evaluate the cost-effectiveness of different screening patterns for active chronic hepatitis C virus (HCV) infections utilizing the hepatitis C core antigen test compared to standard care in the context of a general screening program in a high-prevalence country.

Methods: This study developed a decision analytic model to estimate the cost-effectiveness of four screening algorithms for the detection of active HCV infections among asymptomatic individuals with an unknown HCV status in a context of high (>5%) HCV prevalence. Three algorithms started with a serological test for antibodies (AB) followed by a nucleic acid test for HCV-RNA (RNA), the HCVAg (AG) assay, or both. An additional single marker screening strategy with AG was added to the analysis. By the example of the Republic of Georgia, strategies were compared in terms of total costs for screening and diagnosis of an active infection from a health system perspective.

Results: Replacing RNA with AG for confirmation of positive AB identified fewer active infections (–110 per 100,000 screened subjects) at significantly reduced total costs (–$2.74 per screened) and costs per diagnosed infection (–$44). Adding a subsequent RNA confirmatory test on AG negative results captured at least the same rate compared to the standard (AB followed by RNA) at still reduced costs (–$1.16 per subject screened, –$22 per case detected). Utilizing AG as the frontline test revealed the highest detection rate (97.9%) at the highest costs (+$3.80 per subject, +$323 per case detected vs standard).

Conclusion: A combined pattern of HCV AB screening followed by sequential confirmation with AG and RNA on AG negatives would provide equal or better diagnostic performance at lower cost over a broad range of scenarios. Potential long-term consequences of screening strategies to patients and society have to be considered, since the latency period for HCV to develop into severe liver disease is long.     

Benefit and Distance Functions

We explore the relationship between R. W. Shephard's input distance function (“Cost and Production Functions,” Princeton Univ. Press, Princeton, 1953) and D. G. Luenberger's benefit function (J. Math. Econ.21(1992a), 461–481). We point out that the latter can be recognized in a production context as a directional input distance function which can exhaustively characterize technologies in both price and input space. D. McFadden's (Cost, revenue, and profit functions,in“Production Economics: A Dual Approach to Theory and Applications, “North-Holland/Elsevier, New York, 1978) composition rules for input sets and input distance functions are then extended to the directional input distance function.Journal of Economic LiteratureClassification Numbers : D21, D24, D29.     

Sustainability of external debt: further evidence from non-linear framework

This study investigates sustainability of external debt under a two-step non-linear framework. The first step uses a general linearity test proposed by Harvey and Leybourne (2007 Harvey, David I. and Leybourne, Stephen J. 2007. Testing for time series linearity. Econometric Journal, 10: 149–165. [Crossref], [Web of Science ®] , [Google Scholar]) to determine the linearity property of external debt. The second step applies a non-linear ADF unit root test proposed by Kapetanios, Shin, and Snell (2003 Kapetanios, G., Y. Shin, and A. Snell. 2003. Testing for a unit root in the nonlinear STAR. Journal of Econometrics 112: 359–79.  [Google Scholar]) on the non-liner processes and the linear ADF test on the linear processes to examine the sustainability of external debt. The analysis of 36 debt and 55 current account ratios identifies strong evidence of non-linearity and sustainability. The results indicate superior performance of the non-linear unit root test over the ADF test in determining the stationary property of the data.     

Calendar effects in structural time series models with trend and season

Calendar effects are analysed in the class of structural time series models one of the two main model based approaches in time series decomposition. While Bell and Hillmer (1983) modeled calendar variation in the ARIMA model based approach, we represent structural models in the generalized regression form which allows to apply classical estimation and test procedures. It turns out that the expected high computaional complexity 0(T ³) in the generalized regression model can be reduced to 0(T). As all parameters are estimated by maximizing the likelihood the Likelihood Ratio statistics can be used to test effects of the calendar composition.     

Nonlinear Mean Reversion across National Stock Markets: Evidence from Emerging Asian Markets

This paper seeks empirical evidence of nonlinear mean-reversion in relative national stock price indices for Emerging Asian countries. It is well known that conventional linear unit root tests suffer from low power against the stationary nonlinear alternative. Implementing the nonlinear unit root tests proposed by Kapetanios et al. (2003 Kapetanios, G., Shin, Y. and Snell, A. 2003. Testing for a unit root in the nonlinear STAR framework. Journal of Econometrics, 112: 359–379. [Crossref], [Web of Science ®] , [Google Scholar]) and Cerrato et al. (2009 Cerrato, M., de Peretti, C., Larsson, R. and Sarantis, N. 2009. “A nonlinear panel unit root test under cross section dependence”. Working Papers 28, Department of Economics, University of Glasgow [Google Scholar]) for the relative stock prices of Emerging Asian markets, we find strong evidence of nonlinear mean reversion, whereas linear tests fail to reject the unit root null for most cases. We also report some evidence that stock markets in China and Taiwan are highly localized.     

A Copula Nonlinear Granger Causality

We propose a new copula nonlinear Granger causality test that is more robust than the current available linear and nonlinear Granger causality tests when there exists an asymmetric and nonlinear directional dependence. To perform the statistical test of the copula nonlinear causality, the Gaussian Copula Marginal Regression (GCMR) model and copula directional dependence (Kim and Hwang, 2017) are employed in this paper. By using GCMR and two-sample permutation test with rank sum statistic for the copula nonlinear Granger causality, we can confirm that the result of the proposed copula nonlinear Granger causality test is a reliable test through the simulated data and real data both for small and large sample sizes.     

Power issues when testing the Markov switching model with the sup likelihood ratio test using U.S. output

The likelihood ratio (LR) test statistic for the test of a linear AR(1) model against the alternative of a Markov switching model does not possess the standard χ² distribution. Garcia (1998) derives the asymptotic distribution of the Sup LR test statistic under these non-standard conditions allowing the researcher to easily compare the two models. This paper examines the power properties of this test statistic using Monte Carlo experiments calibrated to U.S. output growth data. The results suggest a test of reasonable power. When the experiments are calibrated to annual data, power is 82% at 200 observations. When the experiments are calibrated to quarterly data power is 57% for the same sample size. First Version Received: March 2000/Final Version Received: March 2001     

Using the directional shadow elasticity of substitution to measure the concavity of the cost function

The note develops a test of the concavity of a cost function using the directional shadow elasticity of substitution. The DSES generalizes the shadow elasticity of substitution to arbitrary price changes in the plane tangent to the factor price frontier. The test computes the minimum value of the DSES, which should be non-negative for concavity, and the direction in the price space along which it occurs.     

Benefit Transfer Equivalence Tests with Non-normal Distributions

Equivalence testing reverses traditional null and alternative hypotheses—welfare estimates are assumed different unless tests demonstrate that the difference is smaller than a specified tolerance limit. Within benefit transfer, researchers have universally used the “two one-sided t-test” (TOST) equivalence test, an approach that is invalid for non-normal welfare distributions. This paper proposes an alternative based on the difference between independent empirical distributions, denoted the “two one-sided convolutions” (TOSC) test. The TOSC permits valid inference for non-normal distributions. Empirical assessments show large divergences between TOST and TOSC p-values when distributions are non-normal—demonstrating the likelihood of erroneous inference under the TOST.     

Using the U.S. Test of Financial Literacy in Germany—Adaptation and validation

In this article, the authors present the adaptation and validation processes conducted to render the American Test of Financial Literacy (TFL) suitable for use in Germany (TFL-G). First, they outline the translation procedure followed and the various cultural adjustments made in line with international standards. Next, they present results from the validation of the TFL-G's content and relations between test scores and external variables, including test takers' prior economic education and interest in economic topics. Preliminary analyses of data gathered from expert interviews and cognitive labs, and the results of the first administration to first-year higher education students (N = 1,108) indicate that the TFL-G is a valid instrument to assess young adults' understanding of personal finance in Germany. Perspectives for future research are discussed.     

Are house prices in South Africa really nonstationary? Evidence from SPSM-based panel KSS test with a Fourier function

This study applies the sequential panel selection method (SPSM) to investigate the time-series properties of provincial house prices for entire, large, medium and small middle-segments of South Africa. Quarterly time-series data were collected from nine provinces in South Africa for different house-size categories over the period of 1978.Q1 to 2012.Q4. Whereas other panel-based unit-root tests are joint tests of a unit root for all members of a panel and are incapable of determining the mix of integrated of order zero (I(0)) series and integrated of order one (I(1)) series in a panel setting, the SPSM proposed by Chortareas and Kapetanios (2009) can clearly identify how many and which series in the panel are stationary processes by classifying a whole panel into a group of stationary and nonstationary series. The empirical results from several panel-based, as well as standard pure time-series, unit-root tests, indicate that house prices for the nine provinces studied here are either stationary or nonstationary. However, results from the SPSM using the panel version of the Kapetanios et al. (KSS, 2003) test with a Fourier function unequivocally indicate that house prices are stationary for the nine provinces under study. Our test results have important economic and policy implications for South Africa.     

Structural changes and regional disparity in China's inflation: a revisit

The study re-examined the time series properties and regional disparities in Chinese inflation by extending the work of Chong, Zhang, and Feng (2011 Chong, Tai-Leeung, Terence, Ning Zhang and Feng, Qu. 2011. Structural Changes and Regional Disparity in China's Inflation. Economics Bulletien, 31(1): 572–583.  [Google Scholar]). For this purpose we employed the Lagrange Multiplier (LM) unit root test with one structural break and two structural breaks suggested by Lee and Strazicich (2003 Lee, Junsoo, Mark, C. and Strazicich. 2003. Minimum LM Unit Root Test with Two Structural Breaks. Review of Economics and Statistics, 85(4): 1082–1089. [Crossref], [Web of Science ®] , [Google Scholar], 2004 Lee, J. and Strazicich, M. 2004. Minimum LM unit root test with one structural break. Working Paper 04–17, Boone, North Carolina: Department of Economics, Appalachian State University.  [Google Scholar]) and a recently developed ADF type unit root test with two structural breaks of Narayan and Popp (2010 Narayan, Paresh Kumar and Stephan Popp. 2010. A New Unit root test with Two Structural Breaks in Level and Slope at Unknown Time. Journal of Applied Statistics, 37(9): 1425–1438. [Taylor & Francis Online], [Web of Science ®] , [Google Scholar]). We found that national, urban and rural series of the overall inflation series, clothing, and food, national series of education and residence and the rural series of residence and education are stationary. We also found regional disparity in Chinese inflation, but the disparities existed only in education inflation series.