We propose a hybrid approach for estimating beta that shrinks rolling window estimates toward firm-specific priors motivated by economic theory. Our method yields superior forecasts of beta that have important practical implications. First, unlike standard rolling window betas, hybrid betas carry a significant price of risk in the cross-section even after controlling for characteristics. Second, the hybrid approach offers statistically and economically significant out-of-sample benefits for investors who use factor models to construct optimal portfolios. We show that the hybrid estimator outperforms existing estimators because shrinkage toward a fundamentals-based prior is effective in reducing measurement noise in extreme beta estimates. Received May 17, 2011; accepted October 7, 2015 by Editor Geert Bekaert.

The compensating variation was defined by J. R. Hicks (1942) as the amount one would have to deduct from a person's income to make him just as well off after a change in prices and income as he had been in the initial situation. If the compensating variation is positive, the individual is better off under the new situation. Since the compensating variation furnishes a uniquely defined numerical (cardinal) indicator of welfare improvement, it provides an implicit ranking of alternative prospective situations not only relative to the initial situation, but also relative to each other. In practice this appears to be how this and other tools of cost-benefit analysis are actually used: one is interested in knowing not only whether a particular bridge, or a particular excise tax, will lead to an improvement in welfare, but which out of a set of alternative bridges or alternative tax systems will improve welfare the most.1 In this paper we analyze conditions under which the compensating variation can be validly used in this generalized sense; these turn out to be precisely the same as conditions previously derived (see our 1976 paper) for the valid use of consumer' s surplus as a welfare measure. In our final section we analyze the problem of deriving the generalized equivalent and compensating variations (either exactly or approximately) from observable demand functions, by means of generalizations of, or alternatives to, consumer's surplus.

We examined downside and upside risk spillovers from exchange rates to stock prices and vice versa for a set of emerging economies. We characterized the dependence structure between currency and stock returns using copulas and computed downside and upside value-at-risk and conditional value-at-risk. We documented a positive relationship between stock prices and currency values in emerging economies with respect to the US dollar and the euro, with downside and upside spillover risk effects transmitted both ways. Finally, we also documented asymmetries in upside and downside risk spillovers and asymmetric differences in the size of risk spillovers when the domestic currency values against the US dollar and the euro. Our results, consistent with flight-to-quality phenomena, have implications for downside and upside risk management of international investor portfolios in emerging markets.

Many commodity markets contain a strong seasonal component not only at the price level, but also in volatility. In this paper, the importance of seasonal behavior in the volatility for the pricing of commodity options is analyzed. We propose a seasonally varying long-run mean variance process that is capable of capturing empirically observed patterns. Semi-closed-form option valuation formulas are derived. We then empirically study the impact of the proposed Seasonal Stochastic Volatility Model on the pricing accuracy of natural gas futures options traded at the New York Mercantile Exchange (NYMEX) and corn futures options traded at the Chicago Board of Trade (CBOT). Our results demonstrate that allowing stochastic volatility to fluctuate seasonally significantly reduces pricing errors for these contracts.

ABSTRACT A significant proportion of the debt issued by investment‐grade firms has maturities greater than 20 years. In this paper we provide evidence that gap‐filling behavior is an important determinant of these very long‐term issues. Using data on individual corporate debt issues between 1987 and 2009, we find that gap‐filling behavior is more prominent in the very long end of the maturity spectrum where the required risk capital makes arbitrage costly. In addition, changes in the supply of long‐term government bonds affect not just the choice of maturity but also the overall level of corporate borrowing.

This paper provides a novel mechanism for identifying and estimating latent group structures in panel data using penalized techniques. We consider both linear and nonlinear models where the regression coefficients are heterogeneous across groups but homogeneous within a group and the group membership is unknown. Two approaches are considered—penalized profile likelihood (PPL) estimation for the general nonlinear models without endogenous regressors, and penalized GMM (PGMM) estimation for linear models with endogeneity. In both cases, we develop a new variant of Lasso called classifier‐Lasso (C‐Lasso) that serves to shrink individual coefficients to the unknown group‐specific coefficients. C‐Lasso achieves simultaneous classification and consistent estimation in a single step and the classification exhibits the desirable property of uniform consistency. For PPL estimation, C‐Lasso also achieves the oracle property so that group‐specific parameter estimators are asymptotically equivalent to infeasible estimators that use individual group identity information. For PGMM estimation, the oracle property of C‐Lasso is preserved in some special cases. Simulations demonstrate good finite‐sample performance of the approach in both classification and estimation. Empirical applications to both linear and nonlinear models are presented.

Whether, and how, the introduction of deposit insurance affects non-bank financial market development depends on whether banks and non-bank financial markets are substitutes or complements and theory has conflicting views. Using data on 134 countries over a 28-year period and several identification strategies we find that the introduction of deposit insurance retards the equity market, the non-bank depositaries sector, and the banking sector when law and order is weak. While strong law and order mitigates this effect, it does not lead to a positive outcome for all markets. For non-bank financial markets, the effect is greater in the long run so that while deposit insurance increases banking sector development in the long run, it retards non-bank financial markets regardless of the level of law and order. Finally, several design features exacerbate the negative outcomes. Our results have important policy implications for implementing or altering deposit insurance schemes.

DeRosa and Goldstein (hereafter D-G) argue that the appropriate test for a pricing should regard only the sign of a05, the estimated coefficient of the cost change dummy variable (V1) and the sign and magnitude of a7, the estimated coefficient on the interaction term between the cost change dummy variable and the change in variable cost (V1 X AVC). In their words, Since a0 [the intercept term] in equation (2) is the estimated time trend of prices and cannot be uniquely identified with the hypothesis under investigation, it should not be used as evidence of an asymmetry (p. 880, fn. 6). We would argue instead that the magnitude of the intercept term in the price equation lies at the heart of the issue. The intercept term is the empirical manifestation of inflationary inertia, reflecting the expectations of price and wage setters. George Perry has called this phenomenon the inflationary norm. As discussed below, the estimated intercept terms in both crosssectional and time-series price equation studies have been positive and of relatively large magnitude, especially in the 1970's. (D-G's estimated intercept terms are 1.20 and 2.09 for the 1972-76 period.) The alternative configurations of a0 (the intercept term), and a5 (the coefficient on the cost-change dummy variable) are shown in Figure 1. In panel (a), a0 a ,5 = 0 and there is no asymmetry. In panels (b), (c), and (d), the basic result holds: namely, that a given percentage increase in variable cost produces a larger price increase than the decrease in price produced by a cost decrease of the same magnitude. Only in panel (e) does the work in the opposite direction. Generalizing, given a positive value of a0, the basic pricing result holds for all positive values of a5 and for negative values of a5 provided that j a5j > 2ao. Only when I a5 I > 2ao, a5 < 0, does the work in the opposite direction. This latter condition is met neither in our results nor in those reported by D-G.' Our point of agreement with D-G is in their argument that the negative sign on a5, the coefficient on the cost-change dummy variable, is not by itself the appropriate test for the asymmetry. We disagree with them, however, in the degree of importance to be attached to the intercept term.

In the coming year 1979, is it possible to achieve a 5 percent unemployment rate and keep annual inflation down to 4 percent? To raise this question in terms of a particular econometric model, we ask whether there exist values of the policy instruments which will give rise to solutions of 5 and 4 percent, respectively, for unemployment and inflation. What is the most favorable tradeoff relationship between inflation and unemployment implicit in an econometric model of a national economy? In this paper, we wish to point out that for many econometric models actually in use, the tradeoff relationship is not rigid, but can be shifted toward the origin (but usually not all the way to the origin!) by suitable government policies. Accordingly, we suggest that the tradeoff relationship implicit in an econometric model be defined as the set of points in the unemployment-inflation diagram which cannot be dominated. We will explain the circumstances under which there exists such a southwestern boundary for the points depicting the unemploymentinflation combinations that are achievable according to a given model. We will propose a systematic way to locate points on this boundary and demonstrate that our algorithm works. Stimulated by and based upon A. W. Phillips' original paper on the relation between unemployment and the rate of change of money wage rates, numerous studies have appeared to refine, respecify, and estimate structural equations explaining the rates of change in the wage rates, the price level, unemployment and related variables. It soon became apparent that these studies, though useful, may not be sufficient for ascertaining the tradeoff relationship between unemployment and inflation. If unemployment and inflation are viewed as two of the many endogenous variables which are jointly determined by a system of simultaneous econometric equations, their relationship has to be derived by solving a whole system using alternative values for the policy variables subject to government control. The approach of deriving the unemployment-inflation tradeoff by varying the policy variables and solving for these two endogenous variables in an econometric model has been adopted by