Minimax invariant estimation of a continuous distribution function under entropy loss

For the invariant decision problem of estimating a continuous distribution function F with two entropy loss functions, it is proved that the best invariant estimators d ₀ exist and are the same as the best invariant estimator of a continuous distribution function under the squared error loss function L (F, d)=∫|F (t) −d (t) |² dF (t). They are minimax for any sample size n≥1.     

Bayesian Hypothesis Testing: a Reference Approach

For any probability model M={p(x|θ, ω), θεΘ, ωεΩ} assumed to describe the probabilistic behaviour of data xεX, it is argued that testing whether or not the available data are compatible with the hypothesis H₀={θ=θ₀} is best considered as a formal decision problem on whether to use (a₀), or not to use (a₀), the simpler probability model (or null model) M₀={p(x|θ₀, ω), ωεΩ}, where the loss difference L(a₀, θ, ω) –L(a₀, θ, ω) is proportional to the amount of information δ(θ₀, ω), which would be lost if the simplified model M₀ were used as a proxy for the assumed model M. For any prior distribution π(θ, ω), the appropriate normative solution is obtained by rejecting the null model M₀ whenever the corresponding posterior expectation ∫∫δ(θ₀, θ, ω)π(θ, ω|x)dθdω is sufficiently large. Specification of a subjective prior is always difficult, and often polemical, in scientific communication. Information theory may be used to specify a prior, the reference prior, which only depends on the assumed model M, and mathematically describes a situation where no prior information is available about the quantity of interest. The reference posterior expectation, d(θ₀, x) =∫δπ(δ|x)dδ, of the amount of information δ(θ₀, θ, ω) which could be lost if the null model were used, provides an attractive nonnegative test function, the intrinsic statistic, which is invariant under reparametrization. The intrinsic statistic d(θ₀, x) is measured in units of information, and it is easily calibrated (for any sample size and any dimensionality) in terms of some average log‐likelihood ratios. The corresponding Bayes decision rule, the Bayesian reference criterion (BRC), indicates that the null model M₀ should only be rejected if the posterior expected loss of information from using the simplified model M₀ is too large or, equivalently, if the associated expected average log‐likelihood ratio is large enough. The BRC criterion provides a general reference Bayesian solution to hypothesis testing which does not assume a probability mass concentrated on M₀ and, hence, it is immune to Lindley's paradox. The theory is illustrated within the context of multivariate normal data, where it is shown to avoid Rao's paradox on the inconsistency between univariate and multivariate frequentist hypothesis testing.     

Bounding endogenous regressor coefficients using moment inequalities and generalized instruments

The main approach to deal with regressor endogeneity is instrumental variable estimator (IVE), where an instrumental variable (IV) m is required to be uncorrelated to the regression model error term u (COR(m,u)=0) and correlated to the endogenous regressor. If COR(m,u)≠0 is likely, then m gets discarded. But even when COR(m,u)≠0, often one has a good idea on the sign of COR(m,u). This article shows how to make use of the sign information on COR(m,u) to obtain an one‐sided bound on the endogenous regressor coefficient, calling m a ‘generalized instrument’ or ‘generalized instrumental variable (GIV)’. If there are two GIV's m₁ and m₂, then a two‐sided bound or an improved one‐sided bound can be obtained. Our approach is simple, needing only IVE; no non‐parametrics, nor any ‘tuning constants’. Specifically, the usual IVE is carried out, and the only necessary modification is that the estimate for the endogenous regressor coefficient is interpreted as a lower/upper bound depending on the prior notion on the sign of COR(m,u) and some estimable moment. A real data application is done to Korean household data with two or more children to illustrate our approach for the issue of child quantity–quality trade‐off.     

On a study of the exponential and Poisson characteristics of the Poisson process

Based on the exponential and Poisson characteristics of the Poisson process, in this work we present some characterizations of the Poisson process as a renewal process. More precisely, let γ_t be the residual life at time t of the renewal process A={A(t),t≥0 }, under suitable condition, we prove that if Var(γ_t)=E ² (γ_t),∀t≥0, then A is a Poisson process. Secondly, we show that if Var (A(t)) is proportional to E (A(t)), then A is a Poisson process also, and Var (A(t))=E (A(t)). Received: August 1999     

Evaluation of P(Xt>Yt ) when both Xt and Yt are residual lifetimes of two systems

Let the random variables X and Y denote the lifetimes of two systems. In reliability theory to compare between the lifetimes of X and Y there are several approaches. Among the most popular methods of comparing the lifetimes are to compare the survival functions, the failure rates and the mean residual lifetime functions of X and Y. Assume that both systems are operating at time t > 0. Then the residual lifetimes of them are X_t=X?t | X>t and Y_t=Y?t | Y>t, respectively. In this paper, we introduce, by taking into account the age of systems, a time‐dependent criterion to compare the residual lifetimes of them. In other words, we concentrate on function R(t ):=P(X_t>Y_t) which enables one to obtain, at time t, the probability that the residual lifetime X_t is greater than the residual lifetime Y_t. It is mentioned, in Brown and Rutemiller (IEEE Transactions on Reliability, 22 , 1973) that the probability of type R(t) is important for designing as long‐lived a product as possible. Several properties of R(t) and its connection with well‐known reliability measures are investigated. The estimation of R(t) based on samples from X and Y is also discussed.     

A consistent test for multivariate normality based on the empirical characteristic function

LetX ₁,X ₂, …,X _n be independent identically distributed random vectors in IR ^d ,d ⩾ 1, with sample mean and sample covariance matrixS _n. We present a practicable and consistent test for the composite hypothesisH _d: the law ofX ₁ is a non-degenerate normal distribution, based on a weighted integral of the squared modulus of the difference between the empirical characteristic function of the residualsS _n ^−1/2 (X _j − ) and its pointwise limit exp (−1/2|t|²) underH _d. The limiting null distribution of the test statistic is obtained, and a table with critical values for various choices ofn andd based on extensive simulations is supplied.     

Continua of stochastic dominance relations for unbounded probability distributions

Let P = {F,G,…} be the set of all probability distribution functions with support (0, ∞). An unrestricted stochastic dominance relation>_∞ is defined on P for each real ∞ 1, where F >_∞ G means that ^x_y = 0 (x - y)^{∞ - 1} dG(y) ^x_{n = 0}(x − y)^∞−1 dG(y) for all ∞ 0, with < for some x. These relations are partial orders that increase as ∞ increases with limit relation>_∞. A class U_∞ of utility functions u on (0, ∞) is defined in such a way that F >_∞ G iff udF > udG for all u ε U_∞. The U_∞ decrease as ∞ increases and have a non-empty intersection U_∞. Each u ε U_∞ is an increasing function that has derivatives of all orders that alternate in sign. Criteria which tell when F eventually dominates G in the sense of F >_∞ G are noted. Comparisons with bounded stochastic dominance results are made in several places.     

Measuring Spot Variance Spillovers when (Co)variances are Time‐varying – The Case of Multivariate GARCH Models

We propose global and disaggregated spillover indices that allow us to assess variance and covariance spillovers, locally in time and conditionally on time‐t information. Key to our approach is the vector moving average representation of the half‐vectorized ‘squared’ multivariate GARCH process of the popular BEKK model. In an empirical application to a four‐dimensional system of broad asset classes (equity, fixed income, foreign exchange and commodities), we illustrate the new spillover indices at various levels of (dis)aggregation. Moreover, we demonstrate that they are informative of the value‐at‐risk violations of portfolios composed of the considered asset classes.     

Exponent of Cross‐Sectional Dependence: Estimation and Inference

This paper provides a characterisation of the degree of cross‐sectional dependence in a two dimensional array, {x_it,i = 1,2,...N;t = 1,2,...,T} in terms of the rate at which the variance of the cross‐sectional average of the observed data varies with N. Under certain conditions this is equivalent to the rate at which the largest eigenvalue of the covariance matrix of x _t=(x_1t,x_2t,...,x_Nt)′ rises with N. We represent the degree of cross‐sectional dependence by α, which we refer to as the ‘exponent of cross‐sectional dependence’, and define it by the standard deviation, , where is a simple cross‐sectional average of x_it. We propose bias corrected estimators, derive their asymptotic properties for α > 1/2 and consider a number of extensions. We include a detailed Monte Carlo simulation study supporting the theoretical results. We also provide a number of empirical applications investigating the degree of inter‐linkages of real and financial variables in the global economy. Copyright © 2015 John Wiley & Sons, Ltd.     

On the upper and lower semicontinuity of the Aumann integral

Let (T,τ,μ) be a finite measure space, X be a Banach space, P be a metric space and let L₁(μ,X) denote the space of equivalence classes of X-valued Bochner integrable functions on (T,τ,μ). We show that if φ:T×P→2^X is a set-valued function such that for each fixed pεP, φ(·,p) has a measurable graph and for each fixed tεT, φ(t,·) is either upper or lower semicontinuous then the Aumann integral of φ, i.e.,∫_Tφ(t,p)dμ(t)= {∫_Tx(t)dμ(t):xεS_φ(p)}, where S_φ(p)= {yεL₁(μ,X):y(t)εφ(t,p)μ−a.e.}, is either upper or lower semicontinuous in the variable p as well. Our results generalize those of Aumann (1965, 1976) who has considered the above problem for X=Rⁿ, and they have useful applications in general equilibrium and game theory.     

Erneuerungsprozesse mit Wartungsstrategien

Summary The following renewal process is considrred: given intervals (kt ₁,(k+1)t ₁],k=0, 1, 2, ..., 0<t ₁<, there will be with probabilityp, 0p1, a renewal in each interval at a time selected by random. The costs for each of this renewals are a units, while the costs of the other renewals areb units each. The renewal function and the cost function are derivided and their asymptotic behavior is discussed.     

The average number of events per time unit and its estimation

Let ζ _t be the number of events which will be observed in the time interval [0;t] and define as the average number of events per time unit if this limit exists. In the case of i.i.d. waiting-times between the events,E[ζ _t ] is the renewal function and it follows from well-known results of renewal theory thatA exists and is equal to 1/τ, if τ>0 is the expectation of the waiting-times. This holds true also when τ = ∞.A may be estimate by ζ _t /t or where is the mean of the firstn waiting-timesX ₁,X ₂, ...,X _n . Both estimators converage with probability 1 to 1/τ if theX _i are i.i.d.; but the expectation of may be infinite for alln and also if it is finite, is in general a positively biased estimator ofA. For a stationary renewal process, ζ _t /t is unbiased for eacht; if theX _i are i.i.d. with densityf(x), then ζ _t /t has this property only iff(x) is of the exponential type and only for this type the numbers of events in consecutive time intervals [0,t], [t, 2t], ... are i.i.d. random variables for arbitraryt > 0.     

r-te Momente von Punktprozessen

Zusammenfassung {X (t): tR ⁺} sei ein Punktprozeß,H (x) eine konvexe nicht-negative Funktion. Mit Hilfe der bedingten Wahrscheinlichkeitenp _n (t) für genaun Ereignisse (Punkte) im Zeitpunkt punktt unter der Bedingung, daß im Zeitpunktt mindestens ein Ereignis eintritt, wird eine Beziehung formuliert, die für die Existenz des ErwartungswertesE (H (X (t ₀))) notwendig ist. Hat der Punktprozeß unabhängige Zuwächse, und erfüllt die FunktionH (x) einige weitere Bedingungen, so ist die angegebene Beziehung auch hinreichend für die Existenz dieses Erwartungswertes. Für Punktprozesse mit unabhängigen Zuwächsen ergibt sich als unmittelbare Anwendung dieser Aussagen eine notwendige und hinreichende Bedingung für die Existenz vonE X (t ₀) ^r für reellesr1.

---

Summary Let {X (t): tR ⁺} be a point process andH (x) a convex non-negative function. Using the conditional probabilitiesp _n (t) thatn events (points) occur at timet given that at least one event occurs att a condition is formulated which is necessary for the existence ofE (H (X (t ₀))). This condition is sufficient, too, if the point process has independent increments and the functionH (x) fulfils some further conditions. Using these statements one gets a necessary and sufficient condition for the existence ofE X (t ₀) ^r for realr1.

---

---

Herrn ProfessorWeissinger zum 65. Geburtstag am 12. Mai 1978 gewidmet     

Regression‐Kink Approach for Wage Effect on Male Work Hours

In finding the effect of after‐tax wage rate on work hours, the main difficulty is the endogeneity of after‐tax wage rate that equals ‘one minus average tax rate’ times wage rate. To overcome this endogeneity problem, we take advantage of jumps in the marginal income tax rate, which is a regression discontinuity (RD) idea. This RD, in turn, makes the average income tax rate ‘kink‐continuous’, which is a regression kink (RK) idea. We provide a simple economic model resulting in the RD and RK features, explain how to implement RK in practice, and then apply our methods to Korean male data. Our main RK‐based labour supply elasticity estimate 7.16% turned out to be insignificant with t‐value 1.52, but it is much larger than most estimates in the literature. This may be attributed to, among other things, the facts that the RK instrument is unique, that RK identifies only the local elasticity at the kink point and that RK requires large data as regression derivatives are estimated.     

Response Surface Regressions for Critical Value Bounds and Approximate p‐values in Equilibrium Correction Models1

We consider the popular ‘bounds test’ for the existence of a level relationship in conditional equilibrium correction models. By estimating response surface models based on about 95 billion simulated F‐statistics and 57 billion t‐statistics, we improve upon and substantially extend the set of available critical values, covering the full range of possible sample sizes and lag orders, and allowing for any number of long‐run forcing variables. By computing approximate P‐values, we find that the bounds test can be easily oversized by more than 5 percentage points in small samples when using asymptotic critical values.     

A Survey of L1 Regression

L₁ regularization, or regularization with an L₁ penalty, is a popular idea in statistics and machine learning. This paper reviews the concept and application of L₁ regularization for regression. It is not our aim to present a comprehensive list of the utilities of the L₁ penalty in the regression setting. Rather, we focus on what we believe is the set of most representative uses of this regularization technique, which we describe in some detail. Thus, we deal with a number of L₁‐regularized methods for linear regression, generalized linear models, and time series analysis. Although this review targets practice rather than theory, we do give some theoretical details about L₁‐penalized linear regression, usually referred to as the least absolute shrinkage and selection operator (lasso).     

Non‐parametric bounds on quantiles under monotonicity assumptions: with an application to the Italian education returns

Within the inferential context of predicting a distribution of potential outcomes P[y(t)] under a uniform treatment assignment t ∈ T, this paper deals with partial identification of the α‐quantile of the distribution of interest Q_α[y(t)] under relatively weak and credible monotonicity‐type assumptions on the individual response functions and the population selection process. On the theoretical side, the paper adds to the existing results on non‐parametric bounds on quantiles with no prior information and under monotone treatment response (MTR) by introducing and studying the identifying properties of α‐quantile monotone treatment selection (α‐QMTS), α‐quantile monotone instrumental variables (α‐QMIV) and their combinations. The main result parallels that for the mean; MTR and α‐QMTS aid identification in a complementary fashion, so that combining them greatly increases identification power. The theoretical results are illustrated through an empirical application on the Italian returns to educational qualifications. Bounds on several quantiles of ln(wage) under different qualifications and on quantile treatments effects (QTE) are estimated and compared with parametric quantile regression (α‐QR) and α‐IVQR estimates from the same sample. Remarkably, the α‐QMTS & MTR upper bounds on the α‐QTE of a college degree versus elementary education imply smaller year‐by‐year returns than the corresponding α‐IVQR point estimates. Copyright © 2010 John Wiley & Sons, Ltd.     

On kalman filtering,posterior mode estimation and fisher scoring in dynamic exponential family regression

Summary Dynamic exponential family regression provides a framework for nonlinear regression analysis with time dependent parametersβ ₀,β ₁, …,β _t, …, dimβ _t=p. In addition to the familiar conditionally Gaussian model, it covers e.g. models for categorical or counted responses. Parameters can be estimated by extended Kalman filtering and smoothing. In this paper, further algorithms are presented. They are derived from posterior mode estimation of the whole parameter vector (β′₀, …,β′_t) by Gauss-Newton resp. Fisher scoring iterations. Factorizing the information matrix into block-bidiagonal matrices, algorithms can be given in a forward-backward recursive form where only inverses of “small”p×p-matrices occur. Approximate error covariance matrices are obtained by an inversion formula for the information matrix, which is explicit up top×p-matrices. Heinz Leo Kaufmann, my friend and coauthor for many years, died in a tragical rock climbing accident in August 1989. This paper is dedicated to his memory.     

A mechanism for reducing small-business customer waiting-line dissatisfaction

This paper presents a technique whereby a small business (i.e. a one-cash register operation) can reduce customer waiting-line time dissatisfaction in the purchase stage of the consumer decision process. When the queue length reaches or exceeds critical value N^*, another employee is temporarily transferred to the role of ‘server assistant’ to increase the effective service rate; when the queue length eventually decreases to a second critical value N_*, the server assistant returns to primary duties. An optimal customer-reneging decision model is utilized to model the reneging character of the queue. Simulation experiments confirm key hypotheses concerning the behaviour of the queue and compare the effectiveness of a computed (N^*, N_*) policy with that of alternatives.     

Further Evidence on PPP Adjustment Speeds: the Case of Effective Real Exchange Rates and the EMS*

Two different approaches intend to resolve the ‘puzzling’ slow convergence to purchasing power parity (PPP) reported in the literature [see Rogoff (1996) , Journal of Economic Literature, Vol. 34.] On the one hand, there are models that consider a non‐linear adjustment of real exchange rate to PPP induced by transaction costs. Such costs imply the presence of a certain transaction band where adjustment is too costly to be undertaken. On the other hand, there are models that relax the ‘classical’ PPP assumption of constant equilibrium real exchange rates. A prominent theory put together by Balassa (1964, Journal of Political Economy, Vol. 72) and Samuelson (1964 Review of Economics and Statistics, Vol. 46) , the BS effect, suggests that a non‐constant real exchange rate equilibrium is induced by different productivity growth rates between countries. This paper reconciles those two approaches by considering an exponential smooth transition‐in‐deviation non‐linear adjustment mechanism towards non‐constant equilibrium real exchange rates within the EMS (European Monetary System) and effective rates. The equilibrium is proxied, in a theoretically appealing manner, using deterministic trends and the relative price of non‐tradables to proxy for BS effects. The empirical results provide further support for the hypothesis that real exchange rates are well described by symmetric, nonlinear processes. Furthermore, the half‐life of shocks in such models is found to be dramatically shorter than that obtained in linear models.