Benchmark priors for Bayesian model averaging

In contrast to a posterior analysis given a particular sampling model, posterior model probabilities in the context of model uncertainty are typically rather sensitive to the specification of the prior. In particular, ‘diffuse’ priors on model-specific parameters can lead to quite unexpected consequences. Here we focus on the practically relevant situation where we need to entertain a (large) number of sampling models and we have (or wish to use) little or no subjective prior information. We aim at providing an ‘automatic’ or ‘benchmark’ prior structure that can be used in such cases. We focus on the normal linear regression model with uncertainty in the choice of regressors. We propose a partly non-informative prior structure related to a natural conjugate g-prior specification, where the amount of subjective information requested from the user is limited to the choice of a single scalar hyperparameter g_0j. The consequences of different choices for g_0j are examined. We investigate theoretical properties, such as consistency of the implied Bayesian procedure. Links with classical information criteria are provided. More importantly, we examine the finite sample implications of several choices of g_0j in a simulation study. The use of the MC³ algorithm of Madigan and York (Int. Stat. Rev. 63 (1995) 215), combined with efficient coding in Fortran, makes it feasible to conduct large simulations. In addition to posterior criteria, we shall also compare the predictive performance of different priors. A classic example concerning the economics of crime will also be provided and contrasted with results in the literature. The main findings of the paper will lead us to propose a ‘benchmark’ prior specification in a linear regression context with model uncertainty.     

Stochastic processes of temporary equilibria

It is the aim of the paper to study within the framework of an ‘overlapping generation model’ the evolution of temporary equilibria. At date t, there are ‘newborn’ agents and ‘old’ agents who were born in previous periods; the old agents hold cash balances (fiat money) that they carried over from the previous period. At the beginning of period t, all agents receive a random endowment of consumption goods. Then the agents exchange these endowments and money on spot markets at date t (trading in future markets is not considered). Once a temporary equilibrium is reached, the economy move to the next date. Agents who were born at date t then become old and meet agents born at period t+1.It is shown that the evolution of temporary equilibria in this model leads to analyse the ergodic properties of a certain class of Markov processes with stationary transition probability.     

The Delphi technique as a forecasting tool: issues and analysis

This paper systematically reviews empirical studies looking at the effectiveness of the Delphi technique, and provides a critique of this research. Findings suggest that Delphi groups outperform statistical groups (by 12 studies to two with two ‘ties’) and standard interacting groups (by five studies to one with two ‘ties’), although there is no consistent evidence that the technique outperforms other structured group procedures. However, important differences exist between the typical laboratory version of the technique and the original concept of Delphi, which make generalisations about ‘Delphi’ per se difficult. These differences derive from a lack of control of important group, task, and technique characteristics (such as the relative level of panellist expertise and the nature of feedback used). Indeed, there are theoretical and empirical reasons to believe that a Delphi conducted according to ‘ideal’ specifications might perform better than the standard laboratory interpretations. It is concluded that a different focus of research is required to answer questions on Delphi effectiveness, focusing on an analysis of the process of judgment change within nominal groups.     

Theoretical tests of the rational expectations hypothesis in economic dynamical models

This paper examines different theoretical stability tests of infinite-horizon rational expectations equilibria. These ‘tests’ have different status: two of them express that the considered equilibrium is ‘isolated’ [neither (non-sunspot) equilibria (test 1) nor (well-behaved) sunspot equilibria exist in a neighbourhood (test 2)] and two of them are learning criteria [either standard ‘evolutive learning’ (test 3) or game-theoretical ‘eductive’ learning (test 4)]. Surprisingly, these four tests select the same steady state equilibria in the class of one-dimensional one-step-forward looking economic models. The extension of this equivalence theorem to n-dimensional and then more complex systems is discussed.     

Linear-quadratic approximation, external habit and targeting rules

We examine the linear-quadratic approximation of nonlinear dynamic stochastic optimization problems. A discrete-time version of Magill [1977a. A local analysis of N-sector capital accumulation under uncertainty. Journal of Economic Theory 15(2), 211–219] is generalized to models with forward-looking variables paying special attention to second-order conditions. This is the ‘large distortions’ case in the literature. We apply the approach to monetary policy in a DSGE model with external habit in consumption. We then develop a condition for ‘target-implementability’, a concept related to ‘targeting rules’. Finally, we extend the approach to a comparison between cooperative and non-cooperative equilibria in a two-country model and show that the ‘small distortions’ approximation is inappropriate for this exercise.     

Looking across borders: A test of spatial policy interdependence using local government efficiency ratings

Spatial patterns in (local) government taxation and spending decisions have received a lot of attention. Still, the focus on taxation or expenditure levels in previous studies may be incomplete. Indeed, (rational) individuals are likely to consider the level of spending on (or taxation for) public goods provision simultaneously with how much public goods they actually receive—thus assessing the ‘price/quantity’ of government policies. Therefore, the present paper argues that incumbents may want their ‘price/quantity’ ratio to be close to that in neighbouring regions. Analysing Flemish local governments' efficiency ratings for the year 2000 (which relate total spending to the quantity of locally provided public goods), we confirm the existence of neighbourhood effects in local government policies.     

Comment choosing a monetary instrument the case of supply-side shocks

In a recent paper Craine and Havenner (1981) claim that ‘an additional advantage of the linear-quadratic framework (which is not well known) is that the basic forces affecting the instrument choice decision can be analyzed without an explicit specification for price expectations since the distribution of the error terms in a linear model is independent of the predetermined variables’ (p. 219). The aim of this note is to demonstrate that Craine and Havenner's claim does not generalize to models with expectations of future variables conditioned on present information, such as Turnovsky's (1980) model.     

A theory of formal conceptual definitions: developing theory-building measurement instruments

This study develops the need for formal conceptual definitions (sometimes called nominal definitions) and how to develop better measurement instruments for theory-building. It develops the underlying theory for ‘good’ formal conceptual definitions by defining terms, demonstrating that formal conceptual definitions are needed for all theory-building empirical research, explains how and why ‘good’ formal conceptual definitions are used to develop properties and their measures, and last, it logically explains that good formal conceptual definitions are necessary conditions for construct validity (content validity, criterion validity, convergent validity, and discriminant validity) while statistical tests are sufficient conditions for validity. This theory development explains why formal conceptual definitions are necessary before any traditional statistical empirical validity tests are performed. This study suggests that any statistical validity tests are not meaningful if the concept is not formally defined.In short, the theory of formal conceptual definitions provides a structure to develop ‘good’ measures of the formal theory that leads to ‘good’ empirical theory-building.     

Inside the queue:: hypercongestion and road pricing in a continuous time–continuous place model of traffic congestion

This paper develops a continuous-time–continuous-place dynamic economic model of traffic congestion, based on car-following theory. The model integrates two archetype congestion technologies used in the economics literature: ‘static flow congestion’ and ‘dynamic bottleneck congestion.’ With endogenous departure times and a bottleneck along the route, ‘hypercongestion’ arises as a dynamic equilibrium phenomenon on the upstream road segment. Congestion tolls based on an intuitive dynamic and space-varying generalization of the standard Pigouvian tax rule can hardly be improved upon. A naïve application of a toll schedule based on Vickrey's bottleneck model performs much worse and reduces welfare in the numerical model.     

Age-specific dynamic labor demand and human capital investment

To study the optimal age-specific labor demand and human capital investment at the firm level we extend the standard dynamic labor demand model by introducing ‘age’ as a second dynamic variable and distinguish between two types of workers: ‘low skilled’ and ‘high skilled’. Applying an age-structured optimal control model we derive qualitative features of the optimal age-specific hiring and training effort. For the case of a linear revenue and production function we prove that firms do not anticipate changes in adjustment costs in their optimal decisions. This result no longer holds if a nonlinear revenue or production function is considered.     

Economic dynamics of reservoir sedimentation management: Optimal control with singularly perturbed equations of motion

We analyze the economic dynamics of reservoir sedimentation management using the hydrosuction-dredging sediment-removal system. System dynamics depend on two interdependent hydraulic processes evolving at different rates. The accumulation of water impounded in the reservoir evolves on a ‘fast’ time scale, while the loss of water storage capacity to trapped sediments evolves on a ‘slow’ time scale. We formulate a multidimensional optimal control problem with singularly perturbed equations of motion to accommodate the disparate time scales. We apply singular perturbation methods to approximate (via polynomial series expansion) a ‘slow’ manifold reducing multi-dimensional solution space to the single-dimensional subspace confining long-term dynamics.     

The power of quality models: The example of the SIQ model for performance excellence

Most contemporary total quality management (TQM) practice is influenced, directly or indirectly, by structured, acontextual and standardized quality models. The present paper focuses on the strategic introduction of one such model, namely the Swedish Institute for Quality (SIQ) model for performance excellence, in a Swedish public-sector organization, which we refer to as ‘the Authority.’ We take our theoretical stance from Foucault's concept of ‘power/knowledge.’ In describing the case, we focus on the management team of one of the Authority's ten regions. Our analysis shows the members of the management team using the SIQ model to objectify both the organization and themselves as managers. However, contrary to many critical or managerial accounts, the SIQ model was not totalizing: management subjectivities changed but were not entirely reconstituted, and some resistance to them was generated by the members of the management team, in their role as professionals.     

Myopic utility functions on sequential economies

In Brown and Lewis (1981) continuity in the Mackey topology of (l^∞, l¹) is related to myopic (or impatient) economic behavior. They also show that finer (locally convex) topologies admit continuous non-myopic utility functions. In that work the space of bounded sequences, l^∞, is interpreted as all time sequences of bounded consumption plans. In Brown (1981) the analysis is extended to study the theory of interest on related sequence spaces.This note applies our simple technique for ‘computing’ Mackey continuity of real-valued functions defined on l^∞. Our first result is motivated by Bewley's (1972, app. II) theorem, but extends it in several important ways (on sequential economies). First, Beweley's examples (specialized to the sequential setting) are all ‘temporally separable’,that is, consumption in one time period does not affect indifference sets in another. We give new explicit examples of Cobb-Douglass-like utility functions and show that the ‘obvious’ infinite-dimensional Cobb-Douglass functions are non-myopic. Known equilibrium theory [from Bewley (1972), but pre-dating him in the sequential case] applies to these new examples. Second, we remove the assumptions of concavity and monotony from the proof of continuity.Our second result shows that some of the ‘stationary’ utility functions studied by Koopmans, Diamond and Williamson (1964) are also myopic in the sense of Brown and Lewis. In general their work is based on the finer uniform topology.Finally, we show how to transform our technique so that it applies to Brown's more general sequential economies. A change of variables transfers our examples to these spaces.     

Restricted houseswapping games

Restricted houseswapping games (RHGs) are a generalization of ‘one-sided matching games’, in which we specify a class II* of ‘allowable’ simple trading cycles. The cores of such games may be empty. Given II*, all possible closed RHGs have non-empty cores of II* is ‘strongly balanced’. Examples include the one-sided matching markets (^{Shapley and Scarf. Journal of Mathematical Economics 1974. 1. 23–37}. ^{Tijs et al., OR Spektrum 1984, 6, 119–123}; ^{Quinzii, International Journal of Game Theory 1984, 13, 41–60}) and the two-sided matching markets (^{Gale and Shapley. American Mathematical Monthly 1962. 69, 9–16}; ^{Shapley and Shubik, International Journal of Game Theory 1972, 1, 111–130}: and ^{Demange and Gale Econometrica 1985, 53, 873–888}).We then consider the subclass of RHGs in which there is no transferable resource. In this case, a weaker condition on II*, called ‘weak balancedness’, is sufficient to guarantee core non-emptiness. In addition, if II* is not weakly balanced, then there exists a preference profile such that the strict core of the resultant game is empty.Several other examples are given of II* that are (a) strongly balanced: (b) weakly balanced but not strongly balanced: and (c) not even weakly balanced.Finally, we discuss the issues of equilibrium definition, existence, and core-equilibrium allocation equivalence in RHGs.     

Control of economic systems under the process of data improvement

The paper deals with optimal control of the systems with a certain part of economic parameters not defined completely, namely, in terms of an interval of uncertainty b ± δ. The important idea of our approach is that optimal control should take into account data improvement. The approach implies the process of introducing new data which make cost estimates more precise. The process is described as a random and exogenous one and by its nature may be named a process with ‘independent decrements’ (of entropy). The stationary model of a dynamic system is developed; the system has the interval of uncertainty as its phase point.     

Dynamic programming for non-additive stochastic objectives

We derive the existence of an optimum and the techniques of dynamic programming for non-additive stochastic objectives. Our key assumption for non-negative objectives is that asymptotic impatience exceeds asymptotic ‘mean’ growth, where ‘mean’ growth is derived not only from intertemporal inelasticity and the random return on investment but also from the curvature of the non-additive stochastic aggregator (i.e. the ‘certainty equivalent’). We provide broad families of new, interesting, and tractable examples. They illustrate that ‘mean’ growth can exist even when the distribution of returns has unbounded support, that power discounting often implies infinite asymptotic impatience, and that non-positive objectives are easily handled with few restrictions on growth.     

Two planar facility location problems with high-speed corridors and continuous demand

The planar minisam (‘median’) and minimax (‘center’) facility location problems are examined under the assumptions that: demand is continuously and uniformly distributed: the L₁ (right-angle) metric is in use; and the planar region is traversed by a high-speed corridor (highway) running parallel to one of the directions of travel. For the minisum problem in a rectangular region, it is shown that, for any combination of problem parameters, there are only two candidate points in the region for the optimum location of the facility. This is also shown to be true for any convex and symmetric planar region. For the minimax problem in a rectangular region, there are only three candidate points for the optimal location. Some extensions and conjectures for the minisum problem involving more than one highway are also discussed.     

Implementing property tax reforms in Bangalore: an actor-network perspective

Many municipal governments in developing countries are experiencing serious difficulties in dealing with the rapid growth of population in cities. Catalysed through processes of globalisation, the growth of cities, especially the so-called ‘mega-cities’, has not been matched by investments in infrastructure like roads and power due to the inability of local government organizations to collect adequate revenues from sources like property tax. Recognizing these difficulties, agencies such as the Asian Development Bank have initiated systems for property tax reforms in a number of Asian cities, including Bangalore. In this paper, we discuss the story of this reform process drawing upon the actor-network perspective on translations. We argue that such a perspective helps to go beyond studying innovation processes through the lens of ‘technology diffusion’, and provides a more interesting and insightful perspective of ‘technology translation’.     

Modeling monetary policy as an unobserved variable

This paper presents a model of the U.S. Monetary Authority. It is assumed that monetary policy is a single dimensioned unobserved variable that links changes in a set of ‘causal’ variables representing economic goals and changes in money market ‘indicator’ variables. Several issues are examined pertaining to the weights attached to the causal variables including: the lag structure, changes over time, and asymmetric effects of positive and negative changes. As a by product of the model, the paper presents an estimated index of monetary policy for the years 1955–1975.     

Management of a capital stock by Strotz's naive planner

The capital management problem posed by R.H. Strotz is analyzed for the case of the ‘naive’ planner who fails to anticipate changes in his own preferences. By imposing progressively stronger restrictions on the primitives of the problem – namely, the discounting function, the utility index function, and the investment technology – the planner's behavior is characterized first as the solution to an ordinary differential equation and then via explicit formulae. Inasmuch as these characterizations leave the discounting function essentially unrestricted, the theory can accommodate, in particular, decision makers who discount time according to the hyperbolic and ‘quasi-hyperbolic’ curves used in applied work and said to be supported by psychological studies. Comparative statics of the model are discussed, as are extensions of the analysis to allow for credit constraints, limited foresight, and partial commitment.