Artificial neural network regression models in a panel setting: Predicting economic growth

Economic time series often feature non-linear structures such as non-linear time trends, non-linear autoregressive effects, and non-linear interaction effects. In this paper, it is shown that artificial neural network regression models are suitable tools for the analysis of economic panel data because they allow for a compromise between the ability to model these features and the model size. As model specification is a concern in artificial neural network models, previous approaches are discussed critically. It is shown that the growth rates of the gross domestic product of 24 industrialized economies in the period 1992–2016 follow a non-linear time trend which cannot be explained by autoregressive features or polynomial time variables. The unrestricted functional form of the time trend in the artificial neural network model is also the main reason for the superior statistical performance compared to conventional panel models. This is confirmed by out-of-sample predictions for 2017.     

Obtaining any Wald statistic you want

While the Wald test can be used to test a non-linear hypothesis in a linear or non-linear regression model it is known that a particular hypothesis can be written in many ways when non-linear forms are permitted. This paper illustrates that it is possible to obtain virtually any value of the Wald statistic at different significance levels. It is also shown that in small samples the use of the χ² or F approximation for the distribution of the Wald statistic can be misleading for some forms of the non-linear Wald test.     

非线性系统的决定论与非决定论

非线性系统的决定论问题是科学领域内的一个十分复杂的问题。非线性科学表现出的许多特征使决定论面临巨大冲击,动摇着经典科学基于简单性信念的世界图景,引发持续不休的论争和困惑。基于当代力学学科所具有的普遍非线性特征,结合当代科技革命所带来的新的变化和新的要求,首先对非线性系统进行了区别于数学形式上的非线性定义,力求突破通常对决定论与非决定论的简单二分。在对决定论与非决定论概念重新界定和分析的基础上,给出了态的哲学定义,进而提出了态决定论的观点,指出：非线性是对严格决定论的彻底否定,但不是对非决定论的肯定,因此,并不必然导致科学的非决定论。态决定论的非线性、外部扰动、多因素耦合、结构和参数不确定性或时变性等多因素作用,使对非线性系统的研究呈现出新的科学图景,     

Uncovered interest parity and risk premium convergence in Central and Eastern European countries

This study applies non-linear threshold unit-root test to investigate the non-stationary properties of the uncovered interest parity (UIP) with risk premium for ten Central and Eastern European (CEE) countries. We find that non-linear threshold unit-root test has higher power than linear method suggested by Caner and Hansen (2001) if the true data generating process of risk premium convergence is in fact a stationary non-liner process. We examine the validity of UIP from the non-linear point of view and provide robust evidence clearly indicating that UIP holds true for seven countries. Our findings point out that capital mobility and exchange market efficiency are in these CEE countries with non-linear way.     

Do consumers benefit from quantity discounts?

A monopoly selling to identical consumers gains at their expense if non-linear pricing is permitted. In contrast, under duopoly non-linear pricing may benefit consumers and lower equilibrium profit.     

Testing for fiscal sustainability: New evidence from the G-7 and some European countries

Whether or not a government deficit is sustainable has important implications for policy. If the debt of a nation is sustainable, then it implies that the government should have no incentive to default on its internal debt. In this article we examine whether or not the debt-GDP ratios of the G-7 and some European countries can be characterized by a unit root process with the non-linear trend and asymmetric adjustment. The econometric methodology allows us to determine whether the stationarity holds for the government's debt–GDP ratio after considering the non-linear trend. Among the main results, it is found that it is very likely that the debt–GDP ratios of Canada, Germany, the US and Italy are stationarity after taking account of the non-linear trend in the long run. Nevertheless, it is model-dependent for the debt–GDP ratios of these countries to be asymmetrically adjusted after taking the non-linear trend into consideration.     

Using Genetic Algorithms to Estimate and Validate Bioeconomic Models: The Case of the Ibero-atlantic Sardine Fishery

Synopsis The Neo-classical approach to fisheries management is based on designing and applying bioeconomic models. Traditionally, the basic bioeconomic models have used pre-established non-linear functional forms (logistic, Cobb–Douglas) in order to try to reflect the dynamics of the renewable resources under study. This assumption might cause misspecification problems and, in consequence, a loss of predictive ability. In this work we intend to verify if there is a bias motivated by employing the said non-linear parametric perspective. For this purpose, we employ a novel non-linear and non-parametric prediction method, called Genetic Algorithms, and we compare its results with those obtained from the traditional methods.     

Real interest rate parity with Flexible Fourier stationary test for Central and Eastern European countries

This study applies stationary test with a Fourier function proposed by Enders and Lee (2012) to test the validity of long-run real interest rate parity (RIRP) to assess the non-stationary properties of the real interest rate convergence for twelve Central and Eastern European (CEE) countries. We find that our approximation has higher power to detect U-shaped breaks and smooth breaks than linear method if the true data generating process of interest rate convergence is in fact a stationary non-linear process. We examine the validity of RIRP from the non-linear point of view and provide robust evidence clearly indicating that RIRP holds true for nine CEE countries. Our findings point out that their interest rate adjustment is mean reversion towards RIRP equilibrium values in a non-linear way.     

Persistent misalignments of the European exchange rates: some evidence from non-linear cointegration

The asymmetric and persistent adjustment of the European real exchange rates is investigated using the framework of non-linear cointegration. The episodes of slow mean-reversion dynamics over the period from 1979 to 1999 are explained. A test of unit root against STAR cointegration is proposed and some complete estimations and stochastic simulations of ESTAR models are presented. The presence of effective non-linear adjustment during the moving of the currencies to their long-run fundamental equilibrium exchange rate value is discussed.     

THE EXTENT OF THE WORLD COFFEE MARKET

This paper employs a novel method to determine whether the prices of various types of coffee are co-integrated. In the spirit of Stigler's arbitrage definition of the market, an attempt is made to test whether all the prices co-integrate in pairs, implying that there is a single market for coffee. This test for co-integration is carried out using both linear and non-linear approaches. This finding demonstrates that the conclusions drawn from the linear and non-linear tests may be drastically different.     

The exact estimation of logit models by non-linear least squares

The purpose of this paper is to examine the relationship between least squares and maximum likelihood estimation, where the likelihood function is the product of two explicit functions. We illustrate the correspondence for the particular case of the logit model, and show that this can be estimated by commonly accessible non-linear least squares estimation packages. Unlike the conventional non-linear least squares approach the estimates obtained following the proposed method are maximum likelihood for all sample sizes.     

The measurement of technical efficiency: a neural network approach

The main purpose of this paper is to provide an introduction to artificial neural networks (ANNs) and to review their applications in efficiency analysis. Finally, a comparison of efficiency techniques in a non-linear production function is carried out. The results suggest that ANNs are a promising alternative to traditional approaches, econometric models and non-parametric methods such as data envelopment analysis, to fit production functions and measure efficiency under non-linear contexts.     

On economic efficiency under non-convexity

This paper investigates economic efficiency under non-convexity. The analysis relies on a generalization of the separating hyperplane theorem under non-convexity. The concept of zero-maximality is used to characterize Pareto efficiency under non-convexity. We show the existence of a separating hypersurface that can be used to provide a dual characterization of efficient allocations. When the separating hypersurface is non-linear, this implies that non-linear pricing is an integral part of economic efficiency. Implications for the decentralization of economic decisions under non-convexity are discussed.     

Does non-linearity help us understand,model and forecast UK stock and bond returns: evidence from the BEYR

The usefulness of non-linear models to provide accurate estimates and forecasts remains an open empirical debate. This paper examines the nature of the estimated relationships and forecasting power of smooth-transition models for UK stock and bond returns using a range of financial and macroeconomic variables as predictors. Notably, evidence of non-linearity is stronger when the bond-equity yield ratio is used as the transition variable. This ratio measures whether stocks are over (under)-valued relative to bonds and can act as a signal for portfolio managers. In-sample results reveal noticeable differences regarding the nature of relationships between the linear and non-linear setting, while results of a recursive forecasting exercise reveal both statistical and economic improvement over a linear model. Overall, these results support the view that non-linear estimates and forecasts can provide useful information for stock market traders, portfolio managers and policy-makers.     

Risk Behavior Under Linear Utility Function

An uncertainly model based on money-market imperfections and on linear utility functions (of money) is reformulated so as to be directly with the traditional non-linear utility approach. We postulate that the utility is a linear functions of “augmented” income. But the augmented income is a non-linear function of the nominal income, due to differential interest rates. This new formulation is shown to generate all the essential implications of the traditional approach as in Friedman and Savage, Markowtiz, Arrow, and Pratt. [020]     

Forecasting Nevada gross gaming revenue and taxable sales using coincident and leading employment indexes

This article provides out-of-sample forecasts of Nevada gross gaming revenue (GGR) and taxable sales using a battery of linear and non-linear forecasting models and univariate and multivariate techniques. The linear models include vector autoregressive and vector error-correction models with and without Bayesian priors. The non-linear models include non-parametric and semi-parametric models, smooth transition autoregressive models, and artificial neural network autoregressive models. In addition to GGR and taxable sales, we employ recently constructed coincident and leading employment indexes for Nevada’s economy. We conclude that the non-linear models generally outperform linear models in forecasting future movements in GGR and taxable sales.     

Structural stability and model design

The paper introduces the concept of structural stability and proposes that it should be considered a necessary property of scientifically valid models. Formalization of the concept is considered in both linear and non-linear models. A strong preference in favour of the wider use of non-linear models is supported by consideration of the dangers of linearization in dynamic models. The importance of structural stability is demonstrated with reference to dynamic rational expectations models which exhibit the saddle-point property. In such cases convergence to equilibrium is shown to be a structurally unstable property which can be forced by restrictive auxillary assumptions, which are highlighted.     

Redistribution with Unobservable Bequests: A Case for Taxing Capital Income

This paper addresses the question of the optimal taxation of labour and interest income in an overlapping generations model with two unobservable characteristics, ability and inheritance. We assume realistically that saving can only be taxed anonymously, whereas the tax on labour earnings can be individualized and made non-linear. In such a setting, we show that a withholding tax on interest income along with a non-linear tax on labour income is desirable. The role of interest income taxation is to indirectly tax inherited wealth.
JEL Classification D 63, H 2     

The process followed by PPP data. On the properties of linearity tests

Recent research has reported the lack of correct size in stationarity test for PPP deviations within a linear framework. However, theoretically well motivated non-linear models, such as the ESTAR, appear to parsimoniously fit the PPP data and provide an explanation for the PPP ‘puzzle’. Employing Monte Carlo experiments the size and power of the non-linear tests are analysed against a variety of nonstationary hypotheses. Aslo the ESTAR model is fitted to data from high inflation economies. The results provide further support for ESTAR specification.