Distribution free estimation of heteroskedastic binary response models using Probit/Logit criterion functions

In this paper estimators for distribution free heteroskedastic binary response models are proposed. The estimation procedures are based on relationships between distribution free models with a conditional median restriction and parametric models (such as Probit/Logit) exhibiting (multiplicative) heteroskedasticity. The first proposed estimator is based on the observational equivalence between the two models, and is a semiparametric sieve estimator (see, e.g. Gallant and Nychka (1987), Ai and Chen (2003) and Chen et al. (2005)) for the regression coefficients, based on maximizing standard Logit/Probit criterion functions, such as NLLS and MLE. This procedure has the advantage that choice probabilities and regression coefficients are estimated simultaneously. The second proposed procedure is based on the equivalence between existing semiparametric estimators for the conditional median model (,  and ) and the standard parametric (Probit/Logit) NLLS estimator. This estimator has the advantage of being implementable with standard software packages such as Stata. Distribution theory is developed for both estimators and a Monte Carlo study indicates they both perform well in finite samples.     

A consistent test for nonlinear out of sample predictive accuracy

In this paper, we draw on both the consistent specification testing and the predictive ability testing literatures and propose an integrated conditional moment type predictive accuracy test that is similar in spirit to that developed by Bierens (J. Econometr. 20 (1982) 105; Econometrica 58 (1990) 1443) and Bierens and Ploberger (Econometrica 65 (1997) 1129). The test is consistent against generic nonlinear alternatives, and is designed for comparing nested models. One important feature of our approach is that the same loss function is used for in-sample estimation and out-of-sample prediction. In this way, we rule out the possibility that the null model can outperform the nesting generic alternative model. It turns out that the limiting distribution of the ICM type test statistic that we propose is a functional of a Gaussian process with a covariance kernel that reflects both the time series structure of the data as well as the contribution of parameter estimation error. As a consequence, critical values that are data dependent and cannot be directly tabulated. One approach in this case is to obtain critical value upper bounds using the approach of Bierens and Ploberger (Econometrica 65 (1997) 1129). Here, we establish the validity of a conditional p-value method for constructing critical values. The method is similar in spirit to that proposed by Hansen (Econometrica 64 (1996) 413) and Inoue (Econometric Theory 17 (2001) 156), although we additionally account for parameter estimation error. In a series of Monte Carlo experiments, the finite sample properties of three variants of the predictive accuracy test are examined. Our findings suggest that all three variants of the test have good finite sample properties when quadratic loss is specified, even for samples as small as 600 observations. However, non-quadratic loss functions such as linex loss require larger sample sizes (of 1000 observations or more) in order to ensure reasonable finite sample performance.     

Efficiency Gains in Rank‐ordered Multinomial Logit Models

This paper considers estimation of discrete choice models when agents report their ranking of the alternatives (or some of them) rather than just the utility maximizing alternative. We investigate the parametric conditional rank‐ordered Logit model. We show that conditions for identification do not change even if we observe ranking. Moreover, we fill a gap in the literature and show analytically and by Monte Carlo simulations that efficiency increases as we use additional information on the ranking.     

Priors and Posterior Computation in Linear Endogenous Variable Models with Imperfect Instruments

In this paper we, like several studies in the recent literature, employ a Bayesian approach to estimation and inference in models with endogeneity concerns by imposing weaker prior assumptions than complete excludability. When allowing for instrument imperfection of this type, the model is only partially identified, and as a consequence standard estimates obtained from the Gibbs simulations can be unacceptably imprecise. We thus describe a substantially improved ‘semi‐analytic’ method for calculating parameter marginal posteriors of interest that only require use of the well‐mixing simulations associated with the identifiable model parameters and the form of the conditional prior. Our methods are also applied in an illustrative application involving the impact of body mass index on earnings. Copyright © 2014 John Wiley & Sons, Ltd.     

Conditional heteroscedasticity of exchange rates: further results based on the fractionally integrated approach

A recent article (Tse, 1998 ) published in this journal analysed the conditional heteroscedasticity of the yen–dollar exchange rate based on the fractionally integrated asymmetric power ARCH model. In this paper, we present replication results using Tse's ( 1998 ) yen–dollar series. We also examine the robustness of Tse's ( 1998 ) findings across different currencies, sample periods and non‐nested GARCH‐type models. Unlike Tse ( 1998 ), we find some evidence of asymmetric conditional volatility for daily returns of currencies measured against the dollar or the yen. Copyright © 2004 John Wiley & Sons, Ltd.     

Testing Steady‐State Restrictions of Linear Rational Expectations Models when Data are Highly Persistent*

Steady‐state restrictions are commonly imposed on highly persistent variables to achieve stationarity prior to confronting rational expectations models with data. However, the resulting steady‐state deviations are often surprisingly persistent indicating that some aspects of the underlying theory may be empirically problematic. This paper discusses how to formulate steady‐state restrictions in rational expectations models with latent forcing variables and test their validity using cointegration techniques. The approach is illustrated by testing steady‐state restrictions for alternative specifications of the New Keynesian model and shown to be able to discriminate between different assumptions on the sources of the permanent shocks.     

Power of Tests for Unit Roots in the Presence of a Linear Trend*

Dickey and Fuller [Econometrica (1981) Vol. 49, pp. 1057–1072] suggested unit‐root tests for an autoregressive model with a linear trend conditional on an initial observation. TPower of tests for unit roots in the presence of a linear trendightly different model with a random initial value in which nuisance parameters can easily be eliminated by an invariant reduction of the model. We show that invariance arguments can also be used when comparing power within a conditional model. In the context of the conditional model, the Dickey–Fuller test is shown to be more stringent than a number of unit‐root tests motivated by models with random initial value. The power of the Dickey–Fuller test can be improved by making assumptions to the initial value. The practitioner therefore has to trade‐off robustness and power, as assumptions about initial values are hard to test, but can give more power.     

A Study of Vertical Integration and Vertical Divestiture: The Case of Store Brand Milk Sourcing in Boston

This article investigates whether a retailer’s store brand supply source impacts vertical pricing and supply channel profitability. Using chain‐level retail scanner data, a random coefficients logit demand model is estimated employing a Bayesian estimation approach. Supply models are specified conditional on demand parameter estimates. Bayesian decision theory is applied to select the best fitting pricing model. Results indicate that a vertically integrated retailer engages in linear pricing for brand manufacturers’ products while competing retailers make nonlinear pricing contracts with brand manufacturers for branded products and store brands. A simulated vertical divestiture based on real world events provides evidence for improved channel efficiency.     

Multi‐state Stochastic Processes: A Statistical Flowgraph Perspective

Two‐state models (working/failed or alive/dead) are widely used in reliability and survival analysis. In contrast, multi‐state stochastic processes provide a richer framework for modeling and analyzing the progression of a process from an initial to a terminal state, allowing incorporation of more details of the process mechanism. We review multi‐state models, focusing on time‐homogeneous semi‐Markov processes (SMPs), and then describe the statistical flowgraph framework, which comprises analysis methods and algorithms for computing quantities of interest such as the distribution of first passage times to a terminal state. These algorithms algebraically combine integral transforms of the waiting time distributions in each state and invert them to get the required results. The estimated transforms may be based on parametric distributions or on empirical distributions of sample transition data, which may be censored. The methods are illustrated with several applications.     

CONDITIONALLY HETEROSKEDASTIC FACTOR MODELS WITH SKEWNESS AND LEVERAGE EFFECTS

Conditional heteroskedasticity, skewness and leverage effects are well‐known features of financial returns. The literature on factor models has often made assumptions that preclude the three effects to occur simultaneously. In this paper I propose a conditionally heteroskedastic factor model that takes into account the presence of both the conditional skewness and leverage effects. This model is specified in terms of conditional moment restrictions and unconditional moment conditions are proposed allowing inference by the generalized method of moments (GMM). The model is also shown to be closed under temporal aggregation. An application to daily excess returns on sectorial indices from the UK stock market provides strong evidence for dynamic conditional skewness and leverage with a sharp efficiency gain resulting from accounting for both effects. The estimated volatilitypersistence from the proposed model is lower than that estimated from models that rule out such effects. I also find that the longer the returns' horizon, the fewer conditionally heteroskedastic factors may be required for suitable modeling and the less strong is the evidence for dynamic leverage. Some of these results are in line with the main findings of Harvey and Siddique ( 1999 ) and Jondeau and Rockinger ( 2003 ), namely that accounting for conditional skewness impacts the persistence in the conditional variance of the return process. Copyright © 2012 John Wiley & Sons, Ltd.     

Likelihood‐Based Inference and Prediction in Spatio‐Temporal Panel Count Models for Urban Crimes

We develop a panel count model with a latent spatio‐temporal heterogeneous state process for monthly severe crimes at the census‐tract level in Pittsburgh, Pennsylvania. Our dataset combines Uniform Crime Reporting data with socio‐economic data. The likelihood is estimated by efficient importance sampling techniques for high‐dimensional spatial models. Estimation results confirm the broken‐windows hypothesis whereby less severe crimes are leading indicators for severe crimes. In addition to ML parameter estimates, we compute several other statistics of interest for law enforcement such as spatio‐temporal elasticities of severe crimes with respect to less severe crimes, out‐of‐sample forecasts, predictive distributions and validation test statistics. Copyright © 2016 John Wiley & Sons, Ltd.     

GENERALIZED AUTOREGRESSIVE SCORE MODELS WITH APPLICATIONS

We propose a class of observation‐driven time series models referred to as generalized autoregressive score (GAS) models. The mechanism to update the parameters over time is the scaled score of the likelihood function. This new approach provides a unified and consistent framework for introducing time‐varying parameters in a wide class of nonlinear models. The GAS model encompasses other well‐known models such as the generalized autoregressive conditional heteroskedasticity, autoregressive conditional duration, autoregressive conditional intensity, and Poisson count models with time‐varying mean. In addition, our approach can lead to new formulations of observation‐driven models. We illustrate our framework by introducing new model specifications for time‐varying copula functions and for multivariate point processes with time‐varying parameters. We study the models in detail and provide simulation and empirical evidence. Copyright © 2012 John Wiley & Sons, Ltd.     

A Meta‐analytic Integration and Test of Psychological Climate Dimensionality

There is no consensus as to what constructs should be considered to be psychological climate. Further, there is no clear taxonomy of psychological climate, although the climate literature suggests that psychological climate perceptions should form higher‐order (i.e., abstract, broad) dimensions. Two meta‐analyses have been conducted to evaluate higher‐order frameworks, but neither study tested the psychometric structure of their hypothesized models. The current meta‐analysis estimated the intercorrelations and criterion‐related validity of 23 psychological climate constructs in order to test the theoretical frameworks proposed by prior meta‐analyses. Confirmatory factor analyses supplemented by a method analogous to item‐to‐scale correlations indicated little empirical support for the a priori frameworks. Results suggest that that psychological climate may be parsimoniously represented by two dimensions. Task climate was most strongly indicated by supervisor goal setting, innovation, and organizational responsiveness. Relational climate was most strongly indicated by work group warmth and social rewards. A path analysis was estimated to test whether job satisfaction partially mediated the association between psychological climate and the outcomes of job performance, turnover intentions, and psychological well‐being. Results provide guidance to researchers and practitioners interested in measuring perceptions of the work environment. © 2014 Wiley Periodicals, Inc.     

MULTIVARIATE VOLATILITY MODELING OF ELECTRICITY FUTURES

We model the dynamic volatility and correlation structure of electricity futures of the European Energy Exchange index. We use a new multiplicative dynamic conditional correlation (mDCC) model to separate long‐run from short‐run components. We allow for smooth changes in the unconditional volatilities and correlations through a multiplicative component that we estimate nonparametrically. For the short‐run dynamics, we use a GJR‐GARCH model for the conditional variances and augmented DCC models for the conditional correlations. We also introduce exogenous variables to account for congestion and delivery date effects in short‐term conditional variances. We find different correlation dynamics for long‐ and short‐term contracts and the new model achieves higher forecasting performance compared \to a standard DCC model. Copyright © 2012 John Wiley & Sons, Ltd.     

Tests of Predictive Ability for Vector Autoregressions Used for Conditional Forecasting

Many forecasts are conditional in nature. For example, a number of central banks routinely report forecasts conditional on particular paths of policy instruments. Even though conditional forecasting is common, there has been little work on methods for evaluating conditional forecasts. This paper provides analytical, Monte Carlo and empirical evidence on tests of predictive ability for conditional forecasts from estimated models. In the empirical analysis, we examine conditional forecasts obtained with a VAR in the variables included in the DSGE model of Smets and Wouters (American Economic Review 2007; 97 : 586–606). Throughout the analysis, we focus on tests of bias, efficiency and equal accuracy applied to conditional forecasts from VAR models. Copyright © 2016 John Wiley & Sons, Ltd.     

Modelling Regime‐Specific Stock Price Volatility*

Single‐state generalized autoregressive conditional heteroscedasticity (GARCH) models identify only one mechanism governing the response of volatility to market shocks, and the conditional higher moments are constant, unless modelled explicitly. So they neither capture state‐dependent behaviour of volatility nor explain why the equity index skew persists into long‐dated options. Markov switching (MS) GARCH models specify several volatility states with endogenous conditional skewness and kurtosis; of these the simplest to estimate is normal mixture (NM) GARCH, which has constant state probabilities. We introduce a state‐dependent leverage effect to NM‐GARCH and thereby explain the observed characteristics of equity index returns and implied volatility skews, without resorting to time‐varying volatility risk premia. An empirical study on European equity indices identifies two‐state asymmetric NM‐GARCH as the best fit of the 15 models considered. During stable markets volatility behaviour is broadly similar across all indices, but the crash probability and the behaviour of returns and volatility during a crash depends on the index. The volatility mean‐reversion and leverage effects during crash markets are quite different from those in the stable regime.     

Comovements and asymmetric tail dependence in state housing prices in the USA: A nonparametric approach

We reexamine the methods used in estimating comovements among US regional home prices and find that there are insufficient moments to ensure a normal limit necessary for employing the quasi‐maximum likelihood estimator. Hence we propose applying the self‐weighted quasi‐maximum exponential likelihood estimator and a bootstrap method to test and account for the asymmetry of comovements as well as different magnitudes across state pairs. Our results reveal interstate asymmetric tail dependence based on observed house price indices rather than residuals from fitting autoregressive–generalized autoregressive conditional heteroskedasticity (AR‐GARCH) models.     

On Testing Conditional Sigma – Convergence*

In a cross‐section where the initial distribution of observations differs from the steady‐state distribution and initial values matter, convergence is best measured in terms of σ‐convergence over a fixed time period. For this setting, we propose a new simple Wald test for conditional σ‐convergence. According to our Monte Carlo simulations, this test performs well and its power is comparable with the available tests of unconditional convergence. We apply two versions of the test to conditional convergence in the size of European manufacturing firms. The null hypothesis of no convergence is rejected for all country groups, most single economies, and for younger firms of our sample of 49,646 firms.     

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 forecast comparison of volatility models: does anything beat a GARCH(1,1)?

We compare 330 ARCH‐type models in terms of their ability to describe the conditional variance. The models are compared out‐of‐sample using DM–$ exchange rate data and IBM return data, where the latter is based on a new data set of realized variance. We find no evidence that a GARCH(1,1) is outperformed by more sophisticated models in our analysis of exchange rates, whereas the GARCH(1,1) is clearly inferior to models that can accommodate a leverage effect in our analysis of IBM returns. The models are compared with the test for superior predictive ability (SPA) and the reality check for data snooping (RC). Our empirical results show that the RC lacks power to an extent that makes it unable to distinguish ‘good’ and ‘bad’ models in our analysis. Copyright © 2005 John Wiley & Sons, Ltd.