David Edgerton | Lund University (original) (raw)

Papers by David Edgerton

RePEc: Research Papers in Economics, Jan 26, 2005

Recent empirical studies suggest that the Fisher hypothesis, stating that inflation and nominal i... more Recent empirical studies suggest that the Fisher hypothesis, stating that inflation and nominal interest rates should cointegrate with a unit parameter on inflation, does not hold, a finding at odds with many theoretical models. This paper argues that these results can be explained in part by the low power inherent in univariate cointegration tests and that the use of panel data should generate more powerful tests. In doing so, we propose two new panel cointegration tests, which are shown by simulation to be more powerful than other existing tests. Applying these tests to a panel of monthly data covering the period 1980:1 to 1999:12 on 14 OECD countries, we find evidence supportive of the Fisher hypothesis.

Advances in Econometrics, 2009

... Solna and Jean-Pierre Fouque Bayesian Econometrics, Edited by Siddhartha Chib, William Grifﬁt... more ... Solna and Jean-Pierre Fouque Bayesian Econometrics, Edited by Siddhartha Chib, William Grifﬁths, Gary Koop and Dek Terrell ... Wei-Shao Wu 1 POTENTIAL BIASES IN SUBSTITUTION ESTIMATES AND VIOLATIONS OF REGULARITY CONDITIONS Leigh Drake and Adrian R ...

Since the beginning of the 1990's, sick leave benefits accruing from the tax-financed Swedish... more Since the beginning of the 1990's, sick leave benefits accruing from the tax-financed Swedish social security insurance have been reduced a number of times. In this paper we use data from the survey on the Household Market and Non-market activities to estimate a fixed effects negative binomial model. We find a direct relationship between the rate of compensation and the amount of sick leave taken

This paper proposes Lagrange multiplier (LM) based tests for the null hypothesis of no cointegrat... more This paper proposes Lagrange multiplier (LM) based tests for the null hypothesis of no cointegration in panel data. The tests are general enough to allow for heteroskedastic and serially correlated errors, individual specific time trends, and a single structural break in both the intercept and slope of each regression, which may be located different dates for different individuals. The limiting distributions of the test statistics are derived, and are found to be standard normal and free of nuisance parameters under the null. In particular, the distributions are found to be invariant not only with respect to trend and structural break, but also with respect to the presence of stochastic regressors. A small Monte Carlo study is also conducted to investigate the small-sample properties of the tests. The results reveal that the tests have small size distortions and good power even in very small samples.

Working Papers Department of Economics Lund University, Feb 27, 2004

Since the beginning of the 1990's, sick leave benefits accruing from the tax-financed Swedish soc... more Since the beginning of the 1990's, sick leave benefits accruing from the tax-financed Swedish social security insurance have been reduced a number of times. In this paper we use data from the survey on the Household Market and Non-market activities to estimate a fixed effects negative binomial model. We find a direct relationship between the rate of compensation and the amount of sick leave taken.

Oxford Bull Econ Stat, 2009

Http Dx Doi Org 10 1080 07474939908800351, Mar 21, 2007

The Breusch-Godfrey test for autocorrelated errors is generalised to cover systems of equations, ... more The Breusch-Godfrey test for autocorrelated errors is generalised to cover systems of equations, and the properties of 18 versions of the test are studied using Monte Carlo methods. We show that only one group of tests regularly has actual size close to the ...

The Barro-Gordon model of monetary policy implies that inflation will inherit the persistency of ... more The Barro-Gordon model of monetary policy implies that inflation will inherit the persistency of the natural rate of unemployment when the central bank cannot commit to a monetary policy rule. The present paper tests the validity of this model in the OECD area by inferring whether unemployment and inflation rates are cointegrated. Evidence from this and other studies suggests that the baseline version of the Barro-Gordon model does not hold, and that this might be due to structural instability induced by changing economic conditions. To be able to test whether this is in fact the case, new panel data cointegration tests are proposed that allow for an arbitrary number of structural breaks in the cointegration relation. Cross-section dependence is also permitted to account for the fact that both unemployment and inflation appear to be affected by shocks that are common across countries. The results based on these new tests suggest that the Barro-Gordon model cannot be rejected. JEL Classification: C12; C32; C33; E30; E52.

Working Papers Department of Economics Lund University, Apr 9, 2004

This paper tests the weak separability of the assets in the Bank of England's household-sector Di... more This paper tests the weak separability of the assets in the Bank of England's household-sector Divisia index from 1977Q1 to 2000Q4. The study is based on a revealed preference framework and uses a nonparametric procedure that jointly tests necessary and sufficient conditions for weak separability, allows for incomplete adjustment of expenditure on monetary services, and allows for measurement errors in the monetary quantity data. The assets included in the Bank of England Divisia index are weakly separable with complete adjustment in two sub-samples covering most of the eighties. A narrower aggregate is weakly separable with complete adjustment in each sub-sample we investigated.

International Economic Review, 1972

IN A RECENT PAPER Goldfeld and Quandt [6] describe two methods of applying Two Stage Least Square... more IN A RECENT PAPER Goldfeld and Quandt [6] describe two methods of applying Two Stage Least Squares (TSLS) to simultaneous models that are nonlinear in the variables. These methods are double regression methods, the initial stage of one (called TSLS1) being ...

Oxford Bulletin of Economics and Statistics, 2008

This paper develops two very simple tests for the null hypothesis of no cointegration in panel da... more This paper develops two very simple tests for the null hypothesis of no cointegration in panel data. The tests are general enough to allow for heteroskedastic and serially correlated errors, unit speciflc time trends, cross-sectional dependence and an unknown structural break in both the intercept and slope of the cointegrated regression, which may be located at difierent dates for difierent

Http Dx Doi Org 10 1080 00949650214678, Oct 29, 2010

The RESET test for functional misspecification is generalised to cover systems of equations, and ... more The RESET test for functional misspecification is generalised to cover systems of equations, and the properties of 7 versions are studied using Monte Carlo methods. The Rao F -test clearly exhibits the best performance as regards correct size, whilst the commonly used LRT (uncorrected for degrees-of-freedom), and LM and Wald tests (both corrected and uncorrected) behave badly even in single

Essays on Mobility and Social Reproduction, 1978

Encyclopedia of Statistical Science Update Volume 3, 1999

Vi Mot Dom Det Dubbla Framlingsskapet I Sjobo, 1989

Journal of Asian Economics, 2010

In a recent paper present strong evidence suggesting that Chinese regional output is diverging, a... more In a recent paper present strong evidence suggesting that Chinese regional output is diverging, a result that strongly contradicts the goal of China's government of a balanced wealth-creation across regions. This paper provides an in-depth analysis of the reasoning behind this finding. Our main result is that the divergence does exist, even when new data and more advanced methods of analysis are used. We also find that it has both an idiosyncratic and a common component. Hence, the increased output inequalities observed at the regional level is due to both region-specific disparities and to disparities between clubs of regions. JEL Classification: C33; O53; R11.

The Fix Point Approach to Interdependent Systems, 1981