Equilibrium Growth and Nonlinear Dynamics in Continuous-Time Models

Building on recent work on equilibrium growth models with externality, I re-examine earlier results from multi-sector growth models in the continuous-time framework, and study the structural relationship between technology and local dynamics. It is shown that it is very easy to construct large classes of economies that exhibit Hopf bifurcations or indeterminacy with constant returns to scale in multi-sector frameworks.
JEL Classification Numbers: C41, E32, O41.     

Testing Equilibrium Models of Regional Disparities

The large and persistent regional disparities of most European economies have been explained as a disequilibrium phenomenon; convergence between backward and successful regions is slow because the equilibrating forces are weak. Recently, two models have emerged where regional disparities are seen as an equilibrium phenomenon; the amenity model, which assumes that high unemployment and low wages reflect favourable living conditions, and the matching model, which views high unemployment as the result of labour market congestion in declining regions. The paper derives some key empirical implications of the models and examines whether the regional pattern of migration, unemployment and wages in Norway conforms with these implications. The results are supportive of the matching model but not of the amenity model.     

Testing Commitment Models of Monetary Policy in China

In the past decade Chinese inflation was not high on average, but it was quite volatile. Back in the 1980s and 1990s, high inflation was a very real problem. What explains the inflationary dynamics in China? In particular, does monetary policy account for the substantial run-ups of inflation, followed by the equally substantial dis-inflation? In the absence of commitment technologies, the monetary authorities may create surprise inflation to achieve higher growth, while private agents would anticipate that and adjust their decisions accordingly, leading to accelerated inflation without a real impact. Do these types of simple time-inconsistency models of monetary policy explain the dynamic pattern of inflation in China? I show that the long-run and short-run restrictions imposed by discretionary policy, when the time-inconsistent policymaker has a desire to push output above potential, are largely rejected by the data. The estimates of the inflation bias under discretion when the policymaker is asymmetrically averse to recessions are not statistically significant either. The analysis contributes to the understanding of Chinese monetary policy and its inflationary implications and also points to the need of further investigation of inflationary behavior during the economic transition.     

Nonlinear Pricing in General Equilibrium Models with Joint Production

This paper considers general equilibrium models of public utilities which produce either public goods or private goods. In the models, cases of increasing returns are not a priori excluded. The products of the public utilities and their costs are allocated to the consumers according to a rule that is dependent on information communicated to the public utilities. We show that if the public utilities follow a nonlinear pricing rule, the equilibrium allocations are always Pareto-optimal. Moreover, the message space is of finite dimensions.
JEL Classification Numbers: D51, D60, H41, H42.     

Testing Alternative Models Of New Product Diffusion

This paper reports upon the testing of the comparative performance of eight epidemic based diffusion models on data describing the diffusion of camcorders and CD players in the UK and cars in West Germany. Standard epidemic models of new product diffusion are modified to allow for the influence of economic factors and compared to a recent model proposed by Karshenas and Stoneman. The performance of the Karshenas and Stoneman model is good, but so is that of an alternative model in the Marketing literature proposed by Easingwood, Mahajan and Muller.     

Bootstrap inference in econometrics

The astonishing increase in computer performance over the past two decades has made it possible for economists to base many statistical inferences on simulated, or bootstrap, distributions rather than on distributions obtained from asymptotic theory. In this paper, I review some of the basic ideas of bootstrap inference. I discuss Monte Carlo tests, several types of bootstrap test, and bootstrap confidence intervals. Although bootstrapping often works well, it does not do so in every case.
Inférence par la méthode d'auto–amorçage (bootstrap) en économétrie. L'incroyable accroissement dans la puissance des ordinateurs au cours des deux dernières décennies a permis aux économistes de fonder plusieurs inférences sur des distributions simulées, ou obtenues par auto–amorçage, plutôt que sur des distributions obtenues par la théorie aymptotique. Dans ce texte, l'auteur passe en revue quelques–unes des idées de base de l'inférence par la méthode d'auto–amorçage. Le texte discute aussi des tests de Monte Carlo, de divers types de tests et des intervalles de confiance obtenus par la méthode d'auto–amorçage. Même si le processus d'auto–amorçage fonctionne souvent bien, cela n'est pas toujours le cas.     

Bootstrap and Pretest J-Type Non-Nested Tests for Orthogonal Regression Models: Some Monte Carlo Evidence

This paper compares the size and power of two J-type tests for weakly correlated or nearly orthogonal non-nested regression models: a bootstrap and a pretest test. The latter seems to outperform the former in terms of its size characteristics, especially when the alternative model has more non-nested regressors and the orthogonality between the two sets of regressors is severe. The bootstrap test does better in terms of power.

     

Estimating Nonlinear DSGE Models with Moments Based Methods

This article suggests a new approach to approximating moments for nonlinear DSGE models. This approach is fast and sufficiently accurate to estimate nonlinear DSGE models. A small financial DSGE model is repeatedly estimated by several modifications of the suggested approach. Approximations of the moments are close to the results of the large sample Monte Carlo estimation. The quality of parameter estimation using our suggested approach is close to the Central Difference Kalman Filter (CDKF); and our suggested approach is much faster.     

中国实际利率平价的KSS非线性单位根检验

中国一直在进行资本和金融项目的渐进改革,通常描述和刻画这一经济规律变化的是利率平价理论。由于近十几年限制我国利率平价的制度约束条件均得到缓解,所以,本文利用基于ESTAR结构的KSS非线性单位根检验分析法,并连同ADF和PP检验一起对我国实际利率平价进行了实证,检验结果表明,实际利率平价假说成立,并遵循非线性稳态过程,利率的非对称调整导致信贷市场和金融市场的信息不对称。这说明短期内实际利率的调整特征是平滑转移的,在长期内,双边国家均无法实施相对独立的货币政策。     

Bootstrap Methods in Econometrics*

There are many bootstrap methods that can be used for econometric analysis. In certain circumstances, such as regression models with independent and identically distributed error terms, appropriately chosen bootstrap methods generally work very well. However, there are many other cases, such as regression models with dependent errors, in which bootstrap methods do not always work well. This paper discusses a large number of bootstrap methods that can be useful in econometrics. Applications to hypothesis testing are emphasized, and simulation results are presented for a few illustrative cases.     

Modeling Nonlinear Autoregressive Distributed Lag Models: A New Approach

It is a common practice in econometrics that estimation is carried out in terms of the reduced form parameters and the structural form parameters are retrieved using the functional relationship between structural form parameters and the reduced form parameters. The reduced form of many useful economic models is a nonlinear distributed lag model (NLADL) with error structure which may have autocorrelation. In addition, the relationship between the reduced form and the structural parameters is often nonlinear and in a ratio form. In such situations existing sampling theory estimation procedures result in estimators for the structural parameters which do not have finite moments and do not possess optimal sampling properties. As an alternative, we propose a two step Bayesian estimation method. The Bayesian method has great potential and allows us to obtain the posterior probability density functions of all parameters of interest. In particular, its application for the analysis of adaptive expectation partial adjustment models for which the reduced form is a NLADL model, has been found extremely useful. An application to Nerlove’s supply response function supports the proposed methodology.

     

Deriving and Testing Efficient Estimates of WTP Distributions in Destination Choice Models

Estimation of welfare measures is often a dominant driver in the empirical literature on nonmarket valuation. To this end, qualitative choice models based on random utility theory have been widely employed in outdoor recreation studies. A frequent goal of applied studies has been the estimation of welfare changes associated with site attribute changes at recreation sites in order to inform regulatory policy and resource management. We review the evolution of the methodology of random utility theory in this field with a focus on taste heterogeneity models and then focus on the recent proposal of specifying utility in the WTP-space (Train K, Weeks M (2005) Discrete choice models in preference space and willing-to-pay space. In: Scarpa R, Alberini A (eds) Applications of simulation methods in environmental and resource economics, chapter 1. Springer, Dordrecht, pp 1–16). Our empirical application is on outdoor alpine recreation data. We emphasize the efficiency and direct testing that using the maximum simulated likelihood estimator affords to practitioners using the WTP-space approach, and illustrate these with examples.     

Simple Procedures for Testing Autoregressive Versus Moving Average Errors in Regression Models

This paper develops several simple separate (or non-nested) procedures for testing autoregressive versus moving average errors in regression models. These asymptotically valid tests are straightforward to calculate: after estimating both models by maximum likelihood methods, the procedure involves testing the significance of variables added to a linearized version of the null model, the added variables being the predictions, or the residuals from the specified alternative model, or the difference of the predictions of the two models. Some small sample evidence on the properties of the tests is presented, as is an empirical application on the Australian unexpected inflation rate series.
JEL Classification Numbers: C12, C22, C52, E31.     

A Bayesian Approach To Testing for Markov-Switching in Univariate and Dynamic Factor Models

Though Hamilton's (1989) Markov-switching model has been widely estimated in various contexts, formal testing for Markov-switching is not straightforward. Univariate tests in the classical framework by Hansen (1992) and Garcia (1998) do not reject the linear model for GDP. We present Bayesian tests for Markov-switching in both univariate and multivariate settings based on sensitivity of the posterior probability to the prior. We find that evidence for Markov-switching, and thus the business cycle asymmetry, is stronger in a switching version of the dynamic factor model of Stock and Watson (1991) than it is for GDP by itself.     

Efficient Estimation and Testing of Alternative Models of Currency Futures Contracts

An efficient systems approach is used to estimate and test two alternative models regarding the pricing of Australian dollar futures contracts traded on the International Monetary Market of the Chicago Mercantile Exchange. Cointegrating relationships among the Australian dollar spot and futures prices, and the US and Australian risk-free rates of interest, suggest alternative error-correction representations for the cost-of-carry model which, with appropriate zero restrictions, yields the unbiased expectations hypothesis. A structural break in the futures price series permits estimation of appropriate models for the full sample in the presence of the break, for the full sample without explicitly modelling the break, and for two separate sub-samples created by the structural break. The restricted and unrestricted cost-of-carry formulations are estimated for all sample sets, the models obtained are found to be statistically adequate, and the qualitative results are reasonably robust across different sample sets for both models. On the basis of the tests of zero restrictions, the cost-of-carry model is found to be empirically superior to the unbiased expectations hypothesis for the four sample sets considered, regardless of the number of cointegrating relations.     

Testing Prediction Performance of Poverty Models: Empirical Evidence from Uganda

This paper examines the performance of a method of predicting poverty rates. Because most developing countries cannot justify the expense of frequent household budget surveys, additional low‐cost methods have been developed and used. The prediction method is based on a model linking the proportion of poor households to suitable explanatory variables (consumption proxies). These consumption proxies are variables that can be collected at much lower cost through smaller annual surveys. Several applications have shown that such models can produce poverty estimates with confidence intervals of a similar magnitude to the poverty estimates from the household budget surveys. There is, however, limited evidence of how well the methods perform out‐of‐sample. A series of seven household budget surveys conducted in Uganda in the period 1993–2005 allows us to test the prediction performance of the model. We test the poverty models by using data from one survey to predict the proportion of poor households in other surveys, and vice versa. The results are encouraging, as all models predict similar poverty trends. Although in most cases the predictions are precise, sometimes they differ significantly from the poverty level estimated from the survey directly.     

Compensating Variation and Hicksian Choice Probabilities in Random Utility Models that are Nonlinear in Income

In this paper we discuss Hicksian demand and compensating variation in the context of discrete choice. We first derive Hicksian choice probabilities and the distribution of the (random) expenditure function in the general case when the utilities are nonlinear in income. We subsequently derive exact and simple formulae for the expenditure and choice probabilities under price (policy) changes conditional on the initial utility level. This is of particular interest for welfare measurement because it enables the researcher to compute the distribution of compensating variation in a simple way. We also derive formulae for the joint distribution of expenditure, the choice before and after a policy change has been introduced.     

Bootstrap confidence intervals for total value in small area statistics

In small area statistics, many problems deal with the estimation of unknown parameters. This paper will consider interval estimation. Three bootstrap confidence intervals of the total value for the small area are proposed. They are obtained by the percentile method, the t-bootstrap method, and the two-stage t-bootstrap method in the case of application of the count post-stratification estimator for total value. The proposed procedures are illustrated with simulation examples in which the investigated variable has the normal or Poisson distribution in population strata. We do not have to know the population or small area distribution for determining the bootstrap confidence intervals for small area parameters. This is the great advantage of bootstrap methods.     

Neoclassical versus Frontier Production Models? Testing for the Skewness of Regression Residuals*

The empirical literature on production and cost functions is divided into two strands. The neoclassical approach concentrates on model parameters, while the frontier approach decomposes the disturbance term to a symmetric noise term and a positively skewed inefficiency term. We propose a theoretical justification for the skewness of the inefficiency term, arguing that this skewness is the key testable hypothesis of the frontier approach. We propose to test the regression residuals for skewness in order to distinguish the two competing approaches. Our test builds directly upon the asymmetry of regression residuals and does not require any prior distributional assumptions.