A COMPARISON OF CONDITIONAL AND UNCONDITIONAL APPROACHES IN VALUE‐AT‐RISK ESTIMATION*

In this paper, we attempt to find the most important factor causing the differences in the performance of Value‐at‐Risk (VaR) estimation by comparing the performances of conditional and unconditional approaches. For each approach, we use various methods and models with different degrees of flexibility in their distributions including S_U‐normal distribution, which is one of the most flexible distribution functions. Our empirical results underscore the importance of the flexibility‐of‐distribution function in VaR estimation models. Even though it seems to be unclear which approach is better between conditional and unconditional approaches, it seems to be clear that the more flexible distribution we use, the better the performance, regardless of which approach we use.     

DISCRIMINATING SHORT AND LONG MEMORY IN FINITE SAMPLES USING SENSITIVITY ANALYSIS: AN APPLICATION TO GROWTH CONVERGENCE

The standard linear model where u_t is generated from an ARFIMA process, is considered. The sensitivity of the predictor and sensitivity of variance estimates of the linear model to long memory are investigated by constructing the statistical measures B_L/S and D_L/S , respectively. B_L/S and D_L/S is interpreted as a sensitivity measure for the long‐memory process without the short‐memory effects. As an application, the memory characteristics of per capita GDP of 30 countries are investigated from the Maddison GDP dataset. It is found that per‐capita GDP exhibits long memory characteristics, and the long‐run growth estimates are sensitive to the long memory characteristics.     

Moment Estimation of the Probit Model with an Endogenous Continuous Regressor

We propose a generalized method of moments (GMM) estimator with optimal instruments for a probit model that includes a continuous endogenous regressor. This GMM estimator incorporates the probit error and the heteroscedasticity of the error term in the first‐stage equation in order to construct the optimal instruments. The estimator estimates the structural equation and the first‐stage equation jointly and, based on this joint moment condition, is efficient within the class of GMM estimators. To estimate the heteroscedasticity of the error term of the first‐stage equation, we use the k‐nearest neighbour (k‐nn) non‐parametric estimation procedure. Our Monte Carlo simulation shows that in the presence of heteroscedasticity and endogeneity, our GMM estimator outperforms the two‐stage conditional maximum likelihood estimator. Our results suggest that in the presence of heteroscedasticity in the first‐stage equation, the proposed GMM estimator with optimal instruments is a useful option for researchers.     

Topological upper limits of mixed Nash equilibria

Our purpose in this article is to prove that given any integer n ≥ 2 and any non-empty compact Polish spaces S ₁, ..., S _n , if for any u ∈ C( S ₁ × ... × S _n , R) ⁿ , we denote by MNE(u) the set of mixed Nash equilibria of (S ₁, ..., S _n , u), then MNE(u) is a non-empty compact subset of P(S ₁) × ... × P(S _n ) and if u ^k → u in C(S ₁ × ... × S _n , R) ⁿ as k → ∞, then lim sup _{k → ∞} MNE (u ^k ) MNE(u). The author would like to thank the referee for offering critical comments on this paper.     

Balanced Contributions Axiom in the Solution of Cooperative Games

In a gamevin characteristic function form, suppose the Banzhaf value ψ is used to pay a coalitionSalready formed. Then coalitionSno longer receivesv(S); instead it receivesR^ψ(S) = ∑_{i ∈ S}ψ_i(v_s), wherev_Sdenotes the subgame of coalitionS. Surprisingly, the Shapley value of this new game Sh(N, R^ψ) is equal to the Banzhaf value ofv. In this paper we establish a similar result for all values satisfying balanced contributions axiom. Additionally, we introduce player's weights to obtain the corresponding result in the nonsymmetric case.Journal of Economic LiteratureClassification Number: C71     

Estimation of a Stratified Error‐Components Model*

This article shows that the presence of incidental parameters in Baltagi and Griffin's (International Economic Review 29 (1988), 745–53) generalized error‐components model implies there is no guarantee feasible and true GLS estimators have the same asymptotic distribution. The article then considers a related stratified error‐components model that has no incidental parameters. In this model, unobserved heterogeneity occurs through variances changing across strata. The article provides an EM algorithm for calculating estimates of the model's parameters without assuming observations can be classified into strata, derives a bootstrap test for identifying the number of strata, and applies the new methods in an example.     

Value at risk estimation by threshold stochastic volatility model

This article proposes a threshold stochastic volatility model that generates volatility forecasts specifically designed for value at risk (VaR) estimation. The method incorporates extreme downside shocks by modelling left-tail returns separately from other returns. Left-tail returns are generated with a t-distributional process based on the historically observed conditional excess kurtosis. This specification allows VaR estimates to be generated with extreme downside impacts, yet remains empirically widely applicable. This article applies the model to daily returns of seven major stock indices over a 22-year period and compares its forecasts to those of several other forecasting methods. Based on back-testing outcomes and likelihood ratio tests, the new model provides reliable estimates and outperforms others.     

Modeling Disinterest and Dislike: A Bounded Bayesian Mixed Logit Model of the UK Market for GM Food

Mixed logit models represent a powerful discrete choice analytical model but require assumptions about the functional form of the parameter distributions. The use of unbounded distributions, such as the normal distribution, may be regarded as unsuitable where theory indicates that all are negatively affected by increases in an attribute, such as price. Bounded distributions such as the triangular and log-normal are unable to model the case where a section of the population is indifferent towards an attribute, while the remainder are negatively disposed toward it. Train and Sonnier’s bounded mixed logit model accommodates these features and is employed in this paper. A censored normal and Johnson’s S_b distribution are used to model preferences in the UK for food attributes, including price and GM technology. Bi-modal distributions are identified regarding GM food: some are unlikely to ever consume it, some are close to indifference and willing to consume at relatively small discounts while the remainder are fairly unresponsive to further price reductions.     

PREDICTING STOCK RETURNS AND VOLATILITY WITH INVESTOR SENTIMENT INDICES: A RECONSIDERATION USING A NONPARAMETRIC CAUSALITY‐IN‐QUANTILES TEST

Evidence of monthly stock returns predictability based on popular investor sentiment indices, namely S^BW and S^PLS as introduced by Baker and Wurgler (2006, 2007) and Huang et al. (2015) respectively are mixed. While, linear predictive models show that only S^PLS can predict excess stock returns, nonparametric models (which accounts for misspecification of the linear frameworks due to nonlinearity and regime changes) finds no evidence of predictability based on either of these two indices for not only stock returns, but also its volatility. However, in this paper, we show that when we use a more general nonparametric causality‐in‐quantiles model of Balcilar et al., (forthcoming), in fact, both S^BW and S^PLS can predict stock returns and its volatility, with S^PLS being a relatively stronger predictor of excess returns during bear and bull regimes, and S^BW being a relatively powerful predictor of volatility of excess stock returns, barring the median of the conditional distribution.     

Cooperation,stability and social welfare under majority rule

We resolve collective irrationality in the stability set, a solution concept for voting games proposed by Rubinstein (J Econ Theory 23:150–159, 1980) to resolve the well-known paradox of voting in the core. A lack of cooperation within winning coalitions prevents their members from eliminating an alternative globally less preferred to another one when they observe the rational behavior defining the stability set. Allowing for the possibility of accompanying a coalitional vote with a binding solidarity agreement is a natural response to this cooperation failure. This leads to a new standard of behavior and a new solution concept called stability set with binding solidarity agreement or S _c -stability set. It is shown that the S _c -stability set and a newly defined version of the Mas-Colell bargaining set for simple games are not comparable with respect to set inclusion. The S _c -stability set includes the core, is included in the stability set and contains only Pareto-optimal alternatives. When individual preferences are complete linear orders, the S _c -stability set is always non-empty.      

Remarks on the application of statistical games to technological development forecasting

This work presents the probability of determining a quantitative forecast of technological development S(t) defined by a set m of parameters S⁽¹⁾(t),S⁽²⁾(t),…,S^(m)(t), based on statistical game theory. Assuming that the coordinates S⁽ⁱ⁾(t) (i = 1, 2,…,m) of a forecasted vector S(t) are stochastic processes with given probabilistic characteristics, a formula of a function forecasting the value of a coordinate S⁽ⁱ⁾(t) of this vector can be obtained. This formula permits to determine a vector of forecasts _τT(x) of technological development S(t) at a given moment t = τ+T.     

Unmediated communication in games with complete and incomplete information

We study the effects of adding unmediated communication to static, finite games of complete and incomplete information. We characterize S^U(G), the set of outcomes of a game G, that are induced by sequential equilibria of cheap talk extensions. A cheap talk extension of G is an extensive-form game in which players communicate before playing G. A reliable mediator is not available and players exchange private or public messages that do not affect directly their payoffs. We first show that if G is a game of complete information with five or more players and rational parameters, then S^U(G) coincides with the set of correlated equilibria of G. Next, we demonstrate that if G is a game of incomplete information with at least five players, rational parameters and full support (i.e., all profiles of types have positive probability), then S^U(G) is equal to the set of communication equilibria of G.     

The supply side in a post-Keynesian two-sector model

Summary In order to avoid missing the wood for trees a brief summary of the results having been obtained above appears to be appropriate.Firstly, we regard it as a result in itself of the present paper to have presented a two-sector model encapsulating a Kaleckian (and Kaldorian) vision of a capitalist economy, a model in which supply conditions of primary products take up a prominent position.Secondly, it belongs to the main results of the paper that it has laid bare, through the model presented, an astoundingly simple pattern in the way economic key-concepts such as activity, employment and distributive shares are affected by on one hand the demand side (which has so far captured an excessive amount of attention in macroeconomic modelling) and on the other hand the largely neglected supply side of the economy.Thirdly, by means of an arbitrary but not implausible numerical example we have attempted to indicate how changes in activity and distributive shares caused by exogeneous changes on the demand and supply side of the economy, respectively, are in themselves crucially dependent on (the assumptions concerning) the supply elasticity of primary products.The notation applied will be as follows C consumption - C ⁰ autonomous element of the consumption function - I investment (gross) - A autonomous expenditure - S savings (gross) - U stock of the primary product - Q real output (not necessarily real income) (gross) - Y income (gross) - W wage bill - L employment - w money wage rate - p price level - mark-up factor - level parameter of the production function pertaining to the primary sector - a labour-input coefficient of the industrial sector - b raw-material input coefficient of the industrial sector - s _w marginal propensity to save out of wages - s marginal propensity to save out of profits - s (weighted) average ofS _w ands - (unit) raw material costs as a proportion of total (unit) prime costs - share of wages in total income - E _{y, x} partial or total elasticity ofY with respect toX. I am most grateful to Søren Gammelgård, Peter Guldager, Erik Strøjer Madsen, Jørgen Ulff-Møller Nielsen, Kurt Pedersen and an anonymous referee for their valuable suggestions and helpful comments on an earlier draft of this paper.     

Power laws in top wealth distributions: evidence from Canada

This article investigates Pareto power law (PPL) behavior at the top of the Canadian wealth distribution. To this end, Canadian Business data on the wealthiest 100 Canadians for the years 1999–2008 are used. The resulting estimates of the PPL exponent ranged from approximately 1.0 to 1.3 depending on the year of analysis and the estimation method used. These estimates are roughly comparable to those based on Forbes’ list of the wealthiest 400 Americans. Furthermore, whereas modified OLS and maximum likelihood estimates of the power law exponents conform to Zipf’s law, the OLS estimates do not. These results raise some concerns about deducing the magnitudes of and trends in the power law exponents based on a single estimation method and highlight the importance of extensive hypothesis testing for model adequacy. The battery of diagnostic tests pertaining to PPL behavior at the top of the Canadian wealth distribution yields some conflicting results.     

The dynamic effects of general purpose technologies on Schumpeterian growth

General purpose technologies (GPTs) are drastic innovations characterized by pervasiveness in use and innovational complementarities. The dynamic effects of a GPT are analyzed within a quality-ladders model of scale-invariant Schumpeterian growth. The diffusion path of a GPT across a continuum of industries is governed by S-curve dynamics. The model generates a unique, saddle-path long-run equilibrium. Along the transition path, the measure of industries that adopt the new GPT increases, consumption per capita falls, and the interest rate rises. The growth rate of the stock market depends negatively on the rate of GPT diffusion and the magnitude of the GPT-ridden R&D productivity gains; and positively on the rate of population growth. It also follows a U-shaped path during the diffusion process of the new GPT. Finally, the model generates transitional growth cycles of per capita GNP.JEL Classification: E3, O3, O4I would like to thank Elias Dinopoulos for encouragement, and for constructive comments and suggestions. I would also like to thank David Figlio, Douglas Waldo, Steven Slutsky and participants in the 9th Biennial Congress of the International Schumpeter Society for useful discussions and suggestions. Any remaining errors are my own responsibility.     

A non-linear approach for predicting stock returns and volatility with the use of investor sentiment indices

The popular sentiment-based investor index S^BW introduced by Baker and Wurgler (2006, 2007) is shown to have no predictive ability for stock returns. However, Huang et al. (2015) developed a new investor sentiment index, S^PLS, which can predict monthly stock returns based on a linear framework. However, the linear model may lead to misspecification and lack of robustness. We provide statistical evidence that the relationship between stock returns, S^BW and S^PLS is characterized by structural instability and inherent nonlinearity. Given this, using a nonparametric causality approach, we show that neither S^BW nor S^PLS predicts stock market returns or even its volatility, as opposed to previous empirical evidence.     

A non-homogeneous non-uniform influence model of innovation diffusion

A model of innovation diffusion which gives unequal weightage to the adopters of different temporal stages and captures commonly observed ups and downs in new product diffusion is proposed. It is shown that our model possesses features of the existing flexible diffusion models and shows better fit which is indicated by the values of R²_adj, mean absolute deviation, and mean percentage error and estimates a larger market potential, M. It has an interesting feature of conversion factors, first increasing then vanishing, much before we approach market saturation, implying that there is a scope of new thrust in converting remaining potential adopters.     

Dynamic specification of demand functions: A cointegration approach

The single-equation approach has been commonly used in the studies of energy demand. However, as most of the data used in the energy demand model are unlikely to be stationary, this factor has to be taken into account when estimating the demand behavior. To overcome this problem, the authors have applied the cointegration and error-correction models to model Chinese coal consumption data. In order to contrast their performance with such traditional models as Hendry's general-to-specific approach, a forecast error comparison exercise has been conducted. In terms ofex post forecast errors, the Engle-Granger error correction model outperforms other chosen models. By using the Engle-Granger approach, it is possible to obtain important information about the behavior of coal demand in China.     

QUANTILE FORECASTS OF FINANCIAL RETURNS USING REALIZED GARCH MODELS*

This article applies the realized generalized autoregressive conditional heteroskedasticity (GARCH) model, which incorporates the GARCH model with realized volatility, to quantile forecasts of financial returns, such as Value‐at‐Risk and expected shortfall. Student's t‐ and skewed Student's t‐distributions as well as normal distribution are used for the return distribution. The main results for the S&P 500 stock index are: (i) the realized GARCH model with the skewed Student's t‐distribution performs better than that with the normal and Student's t‐distributions and the exponential GARCH model using the daily returns only; and (ii) using the realized kernel to take account of microstructure noise does not improve the performance.     

Structural models, information and inherited restrictions

The derived structural estimates of the system βY = γZ + δU impose identifying restrictions on the reduced form estimates ex post. Some or all of the derived structural estimates are presented as evidence of the model's efficacy. In fact, the reduced form inherits a great deal of information from the structure's restrictions and hypothesized sign patterns, limiting the allowable signs for the reduced form. A method for measuring a structural model's statistical information content is proposed. Further, the paper develops a method for enumerating the allowable reduced form outcomes which can be used to falsify the proposed model independently of significant coefficients found for the structural relations.