Conditional mean functions of non-linear models of US output

We compare a number of models of post War US output growth in terms of the degree and pattern of non-linearity they impart to the conditional mean, where we condition on either the previous period's growth rate, or the previous two periods' growth rates. The conditional means are estimated non-parametrically using a nearest-neighbour technique on data simulated from the models. In this way, we condense the complex, dynamic, responses that may be present in to graphical displays of the implied conditional mean. First version received: Feb. 1999/Final version received: June 2001     

Testing the significance and the non-linearity of the Phillips trade-off in the Euro Area

This paper tests the significance and the non-linearity of the Phillips trade-off in the aggregate Euro Area, in an unobserved components model of stochastic NAIRU and trend output featuring the Phillips equation and the Okun law as main identifying equations, with quarterly data for 1970:I-2002:III. The Phillips curve turns out to be linear and its trade-off statistically significant, while non-linearity shows up in the Okun relation. The results prove to be robust to alternative lag length structures of the model, and to alternative non-linear functional forms. The trend-cycle decompositions estimated with the model capture the main features of the Euro Area’s recent macroeconomic record.First version received: 1 September 2003 / Final version received: June 2004CEMPRE - Centro de Estudos Macroeconómicos e Previsão - is supported by the Fundação para a Ciência e a Tecnologia, Portugal, through the Programa Operacional Ciência, Tecnologia e Inovação (POCTI) of the Quadro Comunitário de Apoio III, which is financed by FEDER and Portuguese funds.We thank comments on earlier versions by Fabio Canova, Miguel St Aubyn, Alvaro Almeida, Pete Richardson, Kevin Ross, and two anonymous referees. We acknowledge James D. Hamilton’s help with the confidence bands. The usual disclaimer applies.     

Bayesian modeling of joint and conditional distributions

In this paper, we study a Bayesian approach to flexible modeling of conditional distributions. The approach uses a flexible model for the joint distribution of the dependent and independent variables and then extracts the conditional distributions of interest from the estimated joint distribution. We use a finite mixture of multivariate normals (FMMN) to estimate the joint distribution. The conditional distributions can then be assessed analytically or through simulations. The discrete variables are handled through the use of latent variables. The estimation procedure employs an MCMC algorithm. We provide a characterization of the Kullback–Leibler closure of FMMN and show that the joint and conditional predictive densities implied by the FMMN model are consistent estimators for a large class of data generating processes with continuous and discrete observables. The method can be used as a robust regression model with discrete and continuous dependent and independent variables and as a Bayesian alternative to semi- and non-parametric models such as quantile and kernel regression. In experiments, the method compares favorably with classical nonparametric and alternative Bayesian methods.     

Measuring the persistence of deviations from purchasing power parity with a fractionally integrated STAR model

Empirically, elements of both fractional long memory and threshold non-linearity are present in the real exchange rates of the G-7 countries against the US, notably in the EU countries. Estimated half lives of deviations from PPP using median unbiased corrections to conventional linear autoregressive models corroborate existing evidence related to the PPP paradox as half lives range from at least four years to an infinite number of years. In contrast, for each EU country, accounting for threshold non-linearity results in estimated half lives that can be less than three years even with the allowance for fractional long memory.     

Using Single-Equation Models of Function-of-Functions Type in Sociological Research

This paper argues that single-equation explanatorymodels of many types of social phenomena should not be built in accordance with establishedsociological ways of thinking. In sociological research, the focus is often on the causal mechanisms behindphenomena, and it is often interesting to use models that show the hierarchical causal structure,that is, how influences are nested in the causal process. I propose such a model with a form that reflectsa two-step structure. According to this model, the dependent factor is a product of independentfactors that are linear functions of variables. The model, which should be used with the factor product inunexpanded form, can be assumed to have wide application. However, the models used insociological research and discussed in textbooks are generally very different. They do not have afunction-of-functions form, but take a form in which variables are directly entered. Furthermore,even if they take interaction into consideration, they are linear in an extended sense because they construeit as one or more terms that are products of single variables. In comparison with the proposed typeof model, these models are technically simpler. However, this paper argues that the proposed typeof model is superior in many contexts because it better reflects the causal process.     

Threshold non-linear dynamics between Hang Seng stock index and futures returns

We test the joint dynamics between the Hong Kong Hang Seng Index futures and the underlying cash index using a Bivariate Threshold AutoRegressive model, which is better able to capture the complex return dynamics evident in financial time series. The results are consistent with a three-regime version of the model, where the lead-lag relation between the index and futures returns is a non-linear threshold-type and the regime switching process depends on the state of the threshold variable. This interaction is symmetric rather than unidirectional, with the strength of the interaction dependent on the regime. These three regimes are also characterised by significant variation in volume, which is consistent with liquidity-induced arbitrage trading.     

Inflation and growth: Explaining a negative effect

The paper presents a monetary model of endogenous growth and specifies an econometric model consistent with it. The economic model suggests a negative inflation-growth effect, and one that is stronger at lower levels of inflation. Empirical evaluation of the model is based on a large panel of OECD and APEC member countries over the years 1961–1997. The hypothesized negative inflation effect is found comprehensively for the OECD countries to be significant and, as in the theory, to increase marginally as the inflation rate falls. For APEC countries, the results from using instrumental variables also show significant evidence of a similar behavior. The nature of the inflation-growth profile and differences in this between the regions are interpreted with the credit production technology of the model in a way not possible with a standard cash-only economy. Research assistance by László Konya, Rezida Zakirova, and Anton Nakov and comments by Michal Kejak, Myles Wallace and Toni Braun are kindly acknowledged, along with comments from the 17^th European Economic Association Meetings, Venice, and the 10^th International Panel Data Conference, Berlin. We also thank the editors and referees for valuable comments, and the first author is grateful to Central European University for research funds.     

Lange's 1938 model: dynamics and the “optimum propensity to consume”

Abstract

Oskar Lange's 1938 article “The Rate of Interest and the Optimum Propensity to Consume” is usually associated with the original IS-LM approach of the late 1930s. However, Lange's article was not only an attempt to illuminate Keynes's main innovations but the first part of a wide project that included the development of a theory of economic evolution. This paper aims at showing that Lange's article can help in illuminating critical aspects of this project: in particular, Lange's idea that a synthesis between Kaldor's and Kalecki's theories and that of Schumpeter, might have been possible and that it represented (in intentions) a “modern” and consistent reconstruction of the Marxist theory of the business cycle. Section 2 clarifies Lange's early reflection on dynamics. Section 3 centres on Lange's 1938 static model and indicates the effects of a change of saving on investment. Section 4 suggests a dynamic reconstruction from which are addressed important arguments raised by Lange in a series of papers written between 1934 and 1942.     

Non-linear Co-integration between Nominal Interest Rates and Inflation: An Examination of the Fisher Hypothesis for Asian Countries

Abstract

This paper examines the long-run relationship between nominal interest rates and inflation for a group of Asian countries over the period February 1973–April 2007. We argue that the empirical failure to find evidence supporting the Fisher effect in previous studies may be attributed to the presence of non-linearities in the long-run relationship between nominal interest rates and inflation. We present evidence that the Fisher relation contains significant logistic smooth transition autoregression (LSTAR)-type non-linearity. This type of non-linearity is consistent with inflation targeting and the opportunistic behavior of policy-makers. Applying a non-linear unit root test to the residuals obtained from the Fisher relation decisively rejects the null hypothesis of a unit root against the alternative of non-linear but globally stationary in all the cases.