Rational belief structures and rational belief equilibria

Summary The paper introduces some simplifying tools and methods for studying Rational Beliefs and for proving existence of Rational Belief Equilibria. We identify a set of stable non-stationary stochastic processes, named SIDS processes. Furthermore we introduce the concept of a Rational Belief Structure, which formulates the Rational Beliefs of the agents as beliefs about the distribution of exogenous variables and the beliefs of other agents. The use of the developed apparatus is demonstrated by showing existence of a set of Rational Belief Equilibria in an Overlapping Generations Model with money and one commodity.The results presented in this paper are taken from my Ph.D. thesis from Stanford University ([13]). The dissertation is devoted to the study of the theory of Rational Beliefs as developed by Mordecai Kurz. I gratefully acknowledge the inspiration obtained from innumerable discussions with him over the years about this subject. His suggestions were instrumental for writing the thesis and this paper. The paper benefited much from a thorough reading by Trinidad Casasus, Mordecai Kurz and Ho-Mou Wu. Financial support from The Academy for Research (Forskerakademiet), Aarhus, Denmark and The University of Copenhagen, during my time as a Ph.D. student and from Danish Social Science Research Council thereafter and in addition from Fondazione Eni Enrico Mattei of Milan, Italy is gratefully acknowledged.     

A Model of Financial Markets with Endogenously Correlated Rational Beliefs

This paper studies how communication amongst agents influences the equilibria of a financial economy. We set up a standard overlapping generations model with assets, while allowing for heterogeneous beliefs. The paper explicitly describes how communication causes the beliefs of the agents to be correlated. In particular, it is shown that communication may generate large fluctuations even if the unconditional probability beliefs themselves are independent. Because of the complex nature of the problem, we use simulations to examine the characteristics of the equilibria Part of the results presented in this paper is based on my Ph.D. thesis at Stanford University. I gratefully acknowledge the inspiration obtained from innumerable discussions with Mordecai Kurz about this subject over the years. Also, I appreciate comments from Kenneth J. Arrow and Peter J. Hammond as well as from the participants of the workshop at Stanford University, University of Tokyo and the 1st Illinois workshop in Economic Theory (University of Illinois at Urbana-Champaign) and the anonymous referee     

Speculative trading with rational beliefs and endogenous uncertainty

Summary. This paper introduces the framework of rational beliefs of Kurz (1994), which makes the assumptions of heterogeneous beliefs of Harrison and Kreps (1978) and Morris (1996) more plausible. Agents hold diverse beliefs that are “rational” in the sense of being compatible with ample observed data. In a non-stationary environment the agents only learn about the stationary measure of observed data, but their beliefs can remain non-stationary and diverse. Speculative trading then stems from disagreements among traders. In a Markovian framework of dividends and beliefs, we obtain analytical results to show how the speculative premium depends on the extent of heterogeneity of beliefs. In addition, we demonstrate that there exists a unique Rational Belief Equilibrium (RBE) generically with endogenous uncertainty (as defined by Kurz and Wu, 1996) and that the RBE price is higher than the rational expectation equilibrium price (REE) under some general conditions Received: March 15, 2001; revised version: April 26, 2002 RID="*" ID="*" We are deeply grateful to Mordecai Kurz for his constant encouragement and inspiring guidance over the years. We wish to express our gratitude to an anonymous referee for the very valuable comments provided. We also thank Kenneth Arrow, Peter Hammond, Roko Aliprantis and Nicholas Yannelis for their helpful suggestions and Academia Sinica and the National Science Council of the R.O.C. for their indispensable support. Correspondence to: H.-M. Wu     

Modelling exchange of probabilistic opinions

Summary. This paper studies how communication or exchange of opinions influences correlation of beliefs. The paper focuses on a situation in which agents communicate with each other infinitely many times without observing data. It is an extension to the ‘Expert Problem’ in Bayesian theory, where the informational flow is asymmetric. Moreover, this paper generalizes the existing literature of communication that employs the common prior assumption (CPA) by allowing for heterogeneous beliefs. Some basic convergence results are shown in contrast with the results obtained under the CPA. Furthermore, several economic implications of the basic results are provided. Received: August 27, 2001; revised version: April 16, 2002 RID="*" ID="*" The results presented in this paper are taken from my Ph.D. thesis at Stanford University. I gratefully acknowledge the inspiration obtained from innumerable discussions with Mordecai Kurz about this subject. Also, I appreciate comments from Kenneth J. Arrow, Peter J. Hammond, Maurizio Motolese, Carsten K. Nielsen, Ho-Mou Wu and the anonymous referee.     

On some topological properties of stable measures

Summary The paper shows that the set of stable probability measures and the set of Rational Beliefs relative to a given stationary measure are closed in the strong topology, but not closed in the topology of weak convergence. However, subsets of the set of stable probability measures which are characterized by uniformity of convergence of the empirical distribution are closed in the topology of weak convergence. It is demonstrated that such subsets exist. In particular, there is an increasing sequence of sets of SIDS measures who's union is the set of all SIDS measures generated by a particular system and such that each subset consists of stable measures. The uniformity requirement has a natural interpretation in terms of plausibility of Rational Beliefs.ncial support from The Carlsberg Foundation, The University of Copenhagen, and Danish Social Science Research Council is gratefully acknowledged. I would like to thank Trinidad Casasus, Mordecai Kurz, and Tiefeng Jiang for many useful suggestions for improving the paper, and Karl Vind for helpful discussions about parts of it.     

Vickrey auctions with reserve pricing

Summary. We generalize the Vickrey auction to allow for reserve pricing in a multi-unit auction with interdependent values. In the Vickrey auction with reserve pricing, the seller determines the quantity to be made available as a function of the bidders' reports of private information, and then efficiently allocates this quantity among the bidders. Truthful bidding is a dominant strategy with private values and an ex post equilibrium with interdependent values. If the auction is followed by resale, then truthful bidding remains an equilibrium in the auction-plus-resale game. In settings with perfect resale, the Vickrey auction with reserve pricing maximizes seller revenues.Received: 31 December 2002, Revised: 5 May 2003, JEL Classification Numbers: D44, C78, D82.Correspondence to: Lawrence M. AusubelThe authors gratefully acknowledge the generous support of National Science Foundation Grants SES-97-31025, SES-01-12906 and IIS-02-05489. We appreciate valuable comments from Ilya Segal. Special thanks go to Mordecai Kurz, who served as Larry's dissertation advisor and who introduced both authors to the economics profession back at IMSSS at Stanford. Congratulations and best wishes are extended to Mordecai and his family on the happy occasion of the publication of Assets, Beliefs, and Equilibria in Economic Dynamics: Essays in Honor of Mordecai Kurz, in which this article also appears.     

Effect of credible quality investment with Bertrand and Cournot competition

We show how credible revelation and ability to commit to quality choice effect equilibrium qualities and welfare when product market is either Bertrand or Cournot competition. We show that results depend on the type of competition but not generally on the cost of quality function. We show that with Bertrand competition, the equilibrium qualities are lower with credible commitment. Competition is moderated and producer surplus is higher and consumer surplus lower. With Cournot competition, higher quality will be better but lower quality will be worse with credible commitment. Consumer surplus is always greater with credible commitment and if cost does not increase too quickly with quality, producer surplus will also increase. Thus credible commitment is a collusive device with Bertrand competition but it can improve social welfare with Cournot competition. Received: February 8, 2000; revised version: February 14, 2002 RID="*" ID="*" The idea of this paper originated in the weekly workshops of Mordecai Kurz at Stanford. I am forever in debted to Mordecai and fellow students – Luis Cabral, Peter DeMarzo, John Hillas, Michihiro Kandori, Steve Langois, Patrick McAllister, Steve Sharpe, Peter Streufert, Steve Turnbull and Gyu-Ho Wang – for their criticism and encouragement. I also benefited from comments from Yi-Heng Chen, Jin-Li Hu, Kala Krishna, Jinji Naoto, Thomas J. Prusa, and Shyh-Fang Ueng at various later stages of this work. Last but not least, I am grateful for the detailed comments of the referee.     

Non-stationary,stable Markov processes on a continuous state space

We extend our previous result on simple stable Markov (SSM) processes to the case where the state space is continuous. As anapplication we show the existence of a competitive general equilibrium of a cobweb model where price volatility is generated both by exogenous shocks and by stochastic, so called generating variables (that may be interpreted as sunspots) that govern the correlation of the rational beliefs of individual agents. I would like to thank Danish Social Science Foundation, The Carlsberg Foundation, Stanford Institute for Theoretical Economics (SITE) and Universita Cattolica for financial help. I would also like to thank Trinidad Casasus, Mordecai Kurz, an anonymous referee as well as participants at SITE, ESAM (2004) and ESEM (2004) for helpful discussions and comments. Peter Harremoes provided me with an illuminating counter example and Hiro Nakata provided many comments that helped improve the exposition of the paper. Part of this work is from a paper previously circulated under the title: “Sunspot rational belief structures: anonymity and endogenous uncertainty”.     

Asset price volatility and trading volume with rational beliefs

Summary. This paper develops a model of speculative trading in a large economy with a continuum of investors. In our model the investors are assumed to have diverse beliefs which are rational in the sense of being compatible with observed data. We demonstrate the existence of price amplification effects and show that the equilibrium prices can be higher or lower than the rational expectation equilibrium price. It is also shown that trading volume is positively related to the directions of price changes. Moreover, we study how asset price volatility and trading volume are influenced by belief structures, short selling constraints and the amount of fund available for investment.Received: 23 January 2003, Revised: 30 April 2003, JEL Classification Numbers: D84, G12.We are grateful to Professors Mordecai Kurz, Kenneth Arrow, Kenneth Judd, Carsten Nielsen, Maurizio Motolese, Mark Garmaise, Jean-Michel Grandmont, Peter Hammond, Karl Shell, Jan Werner and participants of the Society for the Advancement of Economic Theory (SAET) Conference and Stanford Institute of Theoretical Economics (SITE) Conference for many helpful suggestions. Correspondence to: H.-M. Wu     

High equity premia and crash fears - Rational foundations

Summary. We show that in a Lucas endowment economy in which the process for dividends is described by a lattice tree subject to infrequent but observable structural breaks, in equilibrium recursive rational learning may inflate the equity risk premium and reduce the risk-free interest rate for low levels of risk aversion. The key condition for these results to obtain is the presence of sufficient initial pessimism. The relevance of these findings is magnified by the fact that under full information our artificial economy cannot generate asset returns matching the empirical evidence for any positive relative risk aversion.Received: 20 September 2003, Revised: 27 April 2005, JEL Classification Numbers: D83, E44, G12.I am indebted to Mordecai Kurz (the co-Editor), one anonymous Referee, Gabriele Camera, Sadayuki Ono, Olesia Verchenko, and Allan Timmermann for many useful comments. I also think seminar paticipants at the European Economic Association meetings in Stockholm (August 2003) and the Midwest Economics Association in St. Louis (March 2003).     

Construction of a state space for interrelated securities with an application to temporary equilibrium theory

Summary We construct an endogenous state space in an exchange economy with possibly infinite horizon. Every period agents trade securities whose payoffs depend on future dividends and asset prices. We reject the perfect foresight assumption on the ground that agents have not only limited knowledge of other individuals' endowments and preferences, but also limited capacity to compute equilibria. We choose instead absence of arbitrage as the principle which allows agents to determine if a system of future prices is possible. We give an alogrithm to compute the set of nonarbitrage prices every period, with both finite and infinite horizon. We then apply this endogenous structure of uncertainty to an infinite horizon temporary equilibrium model.I would like to thank Professor Donald Brown for his constant help and guidance. I have also greatly benefited from helpful discussions with Professors Jacques Drèze, Bernard Dumas, Mordecai Kurz, Carsten Nielsen, Jan Werner, and Ho-Mou Wu.     

AN EXAMPLE OF AN OPTIMAL FORECAST EXHIBITING DECREASING BIAS WITH INCREASING FORECAST HORIZON

Motivated by a central banker with an inflation target, we show that the optimal forecast bias under non‐quadratic loss functions and non‐normal forecast errors can decrease or initially increase and then decrease with the forecast horizon. We initially proof that, if the variable to forecast can be described by a generalized Rayleigh distribution, its conditional mean does in general not constitute the optimal prediction under a symmetric target zone loss function. Subsequently, we approximate the target zone loss function to show the potential for variation in optimal bias over the forecast horizon.     

Estimation and Testing of Forecast Rationality under Flexible Loss

In situations where a sequence of forecasts is observed, a common strategy is to examine „rationality” conditional on a given loss function. We examine this from a different perspective—supposing that we have a family of loss functions indexed by unknown shape parameters, then given the forecasts can we back out the loss function parameters consistent with the forecasts being rational even when we do not observe the underlying forecasting model? We establish identification of the parameters of a general class of loss functions that nest popular loss functions as special cases and provide estimation methods and asymptotic distributional results for these parameters. This allows us to construct new tests of forecast rationality that allow for asymmetric loss. The methods are applied in an empirical analysis of IMF and OECD forecasts of budget deficits for the G7 countries. We find that allowing for asymmetric loss can significantly change the outcome of empirical tests of forecast rationality.     

Outline of forecast theory using generalized cost functions

The cost functions used to form forecasts in practice may be quite different than the squared costs that is often assumed in forecast theory. The impact on evaluation procedures is determined and simple properties for the derivate of the cost function of the errors are found to provide simple tests of optimality. For a very limited class of situations are forecasts based on conditional means optimal, generally, the econometricians needs to provide the whole conditional predicted distribution. Implications for multi-step forecasts and the combination of forecasts are briefly considered.     

Conditional density forecast of China’s energy demand via QRNN model

We propose a new method for conditional density forecast of China’s energy demand through quantile regression neural network (QRNN). This method has at least two advantages. First, it is flexible to explore the true nonlinearity in the energy demand system via neural network structure. Second, it is able to describe the whole conditional distribution of energy demand via quantile regression. In the empirical study on China’s energy demand, QRNN outperforms several classical methods in terms of forecast accuracy both in-sample and out-of-sample. Considering China’s economic and social environment, we set a scenario for predictors and forecast the conditional density of China’s energy demand from 2015 to 2020. The empirical results show that the conditional density curve moves to right and its dispersion increases over time, which indicates that the energy demand in China will keep growing with an average annual rate of 9.672% and its uncertainty is enlarged with 42.210%.     

Macro foundations of micro-economics

This paper attempts to circumvent the nonsense of the representative agent which arises in macroeconomics. It recognises that macro data are relevant to agents' decisions, and so excess demands should contain macro variables as arguments. The macro variables I consider are the price index, unemployment and GNP. This paper should be regarded as a tentative beginning to make macroeconomic theory literate. Received: September 19, 2001; revised version: July 24, 2002 RID="*" ID="*"I am delighted to contribute to this Festschrift for Mordecai now that he has reached the appropriate age.     

Does inflation uncertainty increase with the level of inflation?

This paper presents evidence that if agents forecast inflation rationally, using an estimate of the reduced form equation which generated the data, then the size of their forecast errors is positively correlated with the level of inflation. Forecast errors are measured first as the residuals from a full sample OLS regression, and secondly from one period ahead, outside sample, forecasts using a regression estimated from only data available at the time of the forecast. Thus, agents who form rational expectations about the variance, as well as the mean, of inflation should form conditional variances dependent on the level of inflation, at the date of the forecast.     

Financial liberalization and short-run housing price dynamics

We develop a continuous time, rational expectations, multi-cohort model of an exchange economy with housing, the purchase of which is subject to a down payment (DP) constraint. The timing of the house purchase decision is a crucial endogenous variable, and four determinants of it are identified – the housing services effect, the interest discounting effect, the consumption smoothing effect, and the rate of price increase effect. Cohort effects, and supply constraints, play crucial roles at the aggregative level. We explore in detail the effects of a discrete financial liberalization, and show that if the liberalization is not announced sufficiently far in advance, housing prices will initially overshoot the new stationary equilibrium, and vice versa. Particular attention is paid to the possibility that for a subset of cohorts along the transition path the DP constraint will not bind. An interesting ‘Prisoners’ Dilemma’ is also identified, and policy implications discussedJEL Classification Numbers: E3, R21Valuable comments and suggestions from Phillip Brock, Ho Kong-Weng, Liu Haoming, David McKenzie, David Miles, Jacques Olivier, Phang Sock-Yong, J. Thampapillai, Ping Wang, Wong Wing-Keung, and Zeng Jinli are gratefully acknowledged. I am also immensely indebted to an anonymous referee, whose incisive, deep and patient comments, on successive drafts, helped greatly to sharpen and improve the paper, as well as to the Editor and the Co-Editor, Professor Mordecai Kurz, for their invaluable advice and encouragement. An earlier version was presented at a Conference in Honour of Ronald McKinnon, held at Stanford University in June 2002     

Floating exchange rates versus a monetary union under rational beliefs: the role of endogenous uncertainty

Using the concept of ex-post optimality, we compare different exchange rate regimes, including floating exchange rates and fixed exchange rates with a Monetary Union in a two country OLG model with stochastic endowments. The emphasis of this comparison is on the welfare consequences of agents having incorrect beliefs. We do not assume that agents can hold any beliefs, but rather that their beliefs are rational that is consistent with the observed empirical behavior of the economy. We study a large set of possible policies, but two of them have our particular interest. The first policy implies devaluations in reaction to a negative shock, while the other implies a fixed exchange rate. These policies have very different consequences. The first will for generic beliefs not result in an ex-post optimal allocation. The other policy is on the other hand always feasible and results in an ex-post optimal allocation. When the two countries form a Monetary Union, the ex-post optimal allocation is also achieved. The meaning of “endogenous uncertainty” as an institutionally induced uncertainty is illustrated. Received: September 1, 2001; revised version: 24 June 2002 RID="*" ID="*" I would like to thank Horace W. Brock, Gianluca Cassese, Paula Orlando, Ho-Mou Wu as well as seminar participants at Copenhagen Business School, ESEM98, Keio University, Kyoto University, Osaka University, SITE (Stanford) and University of Copenhagen for many useful comments on the paper. I am also grateful to Mark J. Garmaise, Takako Fujiwara-Greve, and an anonymous referee for many helpful suggestions for improving the paper. Without the many discussions about Rational Beliefs and related issues I have had with Mordecai Kurz over the years, the research presented here would not have been possible. Financial support from The Carlsberg Foundation, Danish Social Research Council, University of Copenhagen and SITE is gratefully acknowledged.     

Estimating loss functions of experts

We propose a new and simple methodology to estimate the loss function associated with experts’ forecasts. Under the assumption of conditional normality of the data and the forecast distribution, the asymmetry parameter of the lin–lin and linex loss function can easily be estimated using a linear regression. This regression also provides an estimate for potential systematic bias in the forecasts of the experts. The residuals of the regression are the input for a test for the validity of the normality assumption. We apply our approach to a large data set of SKU-level sales forecasts made by experts, and we compare the outcomes with those for statistical model-based forecasts of the same sales data. We find substantial evidence for asymmetry in the loss functions of the experts, with underprediction penalized more than overprediction.