Analysis of treatment response data without the joint distribution of potential outcomes

In this paper we show how it is possible to develop a Bayesian framework for analyzing structural models for treatment response data without the joint distribution of the potential outcomes. That this is possible has not been noticed in the literature. We also discuss the computation of the model marginal likelihood and present recipes for finding relevant treatment effects, averaged over both parameters and covariates. As compared to an approach in which the counterfactuals are part of the prior-posterior analysis (as in the work to date), the approach we suggest is simpler in terms of the required prior inputs, computational burden and extensibility to more complex settings.     

Coalition governments in a model of parliamentary democracy

We analyse a model of coalition government in a parliamentary democracy where parties care both for ideology and perks from office, and examine how the magnitude of this tradeoff affects the nature of coalitions that form. It is shown that equilibrium coalitions can be minimal winning, minority or surplus and they may be ideologically disconnected. The types of coalitions that emerge depend upon the relative importance of rents from office and the distribution of party ideologies. Further, there is a non-monotonic relationship between ideological connectedness of coalitions and rents from office.     

Implications of Nash Bargaining for Horizontal Industry Integration

This article shows how horizontal industry integration can arise from transferable asymmetry of technologies and endowments. The Nash bargaining solution suggests that greater technological diversity among coordinating parties yields greater gains from horizontal integration. The framework fits the case where a firm with a superior technology franchises the technology by horizontal integration. The results appear to fit hog production where integration has been primarily horizontal and, in part, broiler production where integration has been both vertical and horizontal. Specifically, technology has been shared through uniform genetic traits, fine-tuned feed rations, and veterinary services specified in grower contracts.     

Determinants and Relationships in Sectoral Trade: A Bilateral Model for Knitwear Clothing

The generalized aggregated trade models do not capture the industry or product‐specific competitive situation and overgeneralize the bilateral cases. As a result, product‐specific trade determinants at the sectoral or bilateral level cannot be sufficiently drawn from such generalized models. This holds true for knitwear clothing products, an important component of international textile trade. To remedy this, we propose a sector‐specific bilateral model in the context of knitwear clothing exports from India to the United States. This pair of countries is chosen due to unilateral trade flows as well as to underline the contrasting features of developed north versus developing south. The vector autoregression (VAR) model was found more appropriate than other available modeling choices. We used monthly frequency data from January 2006 to December 2012. The traditional determinants such as exchange rate and price competitiveness remain relevant. Chinese competition emerges as a significant determinant, which underlines the relevance of a sector‐specific bilateral trade model. The 2009 recession showed a clear impact, albeit for only a few months. Our model is parsimonious but has more explanatory power than generalized models. Policy researchers may further explore the model for more fine‐tuned policy on sector‐specific factors. © 2016 Wiley Periodicals, Inc.     

Modeling and calculating the effect of treatment at baseline from panel outcomes

We propose and examine a panel data model for isolating the effect of a treatment, taken once at baseline, from outcomes observed over subsequent time periods. In the model, the treatment intake and outcomes are assumed to be correlated, due to unobserved or unmeasured confounders. Intake is partly determined by a set of instrumental variables and the confounding on unobservables is modeled in a flexible way, varying both by time and treatment state. Covariate effects are assumed to be subject-specific and potentially correlated with other covariates. Estimation and inference is by Bayesian methods that are implemented by tuned Markov chain Monte Carlo methods. Because our analysis is based on the framework developed by Chib [2004. Analysis of treatment response data without the joint distribution of counterfactuals. Journal of Econometrics, in press], the modeling and estimation does not involve either the unknowable joint distribution of the potential outcomes or the missing counterfactuals. The problem of model choice through marginal likelihoods and Bayes factors is also considered. The methods are illustrated in simulation experiments and in an application dealing with the effect of participation in high school athletics on future labor market earnings.     

Can Robust Optimization Offer Improved Portfolio Performance? An Empirical Study of Indian market

Journal of Quantitative Economics - The emergence of robust optimization has been driven primarily by the necessity to address the demerits of the Markowitz model. There has been a noteworthy...     

EXTREMIST PLATFORMS: POLITICAL CONSEQUENCES OF PROFIT-SEEKING MEDIA

We analyze how information about candidate quality affects the choice of electoral platforms made by an office-motivated political challenger. The incumbent is of known quality and located at the ideal policy of the voter. The voter cares for both policy and the candidates' quality and can learn about the challenger's quality by buying information. A high-quality challenger then has an incentive to signal her quality by choosing a policy that induces the voter to buy information. We first study the benchmark case in which the information is supplied exogenously, and its quality is independent of the challenger's platform; this yields multiple equilibria and indeterminacy of equilibrium platforms. By contrast, when the information is supplied by a profit-maximizing media outlet, its quality depends on the challenger's platform and we obtain a unique equilibrium platform. In particular, when the incumbent's quality is relatively low, the media coverage rises and the challenger's platform diverges further from the voter's ideal policy as the voter's preference for quality increases.     

A Decision Model for E-commerce-enabled Partial Market Exit

Struggling retail chains often try to recover profitability by closing some of their stores. The challenge in this strategy lies in determining how many stores to close, as store exit has implications for both the customers and the supply chain. After a store closes, its customers are lost forever to the competition, unless there is a surviving open store nearby or an electronic alternative such as an e-store. From the supply chain perspective, after a store closes, its supporting regional distribution center is left with less business, and thus reduced viability. This paper develops a decision support model to study the profitability of alternative retail network structures by varying the proportion of stores that are closed, the average price sensitivity of demand, the price difference between the online store and the traditional retailers, and customer retention rates.     

Analysis of high dimensional multivariate stochastic volatility models

This paper is concerned with the Bayesian estimation and comparison of flexible, high dimensional multivariate time series models with time varying correlations. The model proposed and considered here combines features of the classical factor model with that of the heavy tailed univariate stochastic volatility model. A unified analysis of the model, and its special cases, is developed that encompasses estimation, filtering and model choice. The centerpieces of the estimation algorithm (which relies on MCMC methods) are: (1) a reduced blocking scheme for sampling the free elements of the loading matrix and the factors and (2) a special method for sampling the parameters of the univariate SV process. The resulting algorithm is scalable in terms of series and factors and simulation-efficient. Methods for estimating the log-likelihood function and the filtered values of the time-varying volatilities and correlations are also provided. The performance and effectiveness of the inferential methods are extensively tested using simulated data where models up to 50 dimensions and 688 parameters are fit and studied. The performance of our model, in relation to various multivariate GARCH models, is also evaluated using a real data set of weekly returns on a set of 10 international stock indices. We consider the performance along two dimensions: the ability to correctly estimate the conditional covariance matrix of future returns and the unconditional and conditional coverage of the 5% and 1% value-at-risk (VaR) measures of four pre-defined portfolios.