Deriving the polarity and strength of opinions is an important research topic, attracting significant attention over the last few years. In this work, to measure the strength and polarity of an opinion, we consider the economic context in which the opinion is evaluated, instead of using human annotators or linguistic resources. We rely on the fact that text in on-line systems influences the behavior of humans and this effect can be observed using some easy-to-measure economic variables, such as revenues or product prices. By reversing the logic, we infer the semantic orientation and strength of an opinion by tracing the changes in the associated economic variable. In effect, we use econometrics to identify the “economic value of text ” and assign a “dollar value ” to each opinion phrase, measuring sentiment effectively and without the need for manual labeling. We argue that by interpreting opinions using econometrics, we have the first objective, quantifiable, and contextsensitive evaluation of opinions. We make the discussion concrete by presenting results on the reputation system of Amazon.com. We show that user feedback affects the pricing power of merchants and by measuring their pricing power we can infer the polarity and strength of the underlying feedback postings. 1

Abstract. Most online feedback mechanisms rely on voluntary reporting of privately observed outcomes. This introduces the potential for reporting bias, a situation where traders exhibit different propensities to report different outcome types to the system. Unless properly accounted for, reporting bias may severely distort the distribution of public feedback relative to the underlying distribution of private transaction outcomes and, thus, hamper the reliability of feedback mechanisms. This study offers a method that allows users of feedback mechanisms where both partners of a bilateral exchange are allowed to report their satisfaction to “see through ” the distortions introduced by reporting bias and derive unbiased estimates of the underlying distribution of privately observed outcomes. A key aspect of our method lies in extracting information from the number of transactions where one or both trading partners choose to remain silent. We apply our method to a large data set of eBay feedback. Our results confirm the widespread belief that eBay traders are more likely to post feedback when satisfied than when dissatisfied. Furthermore, we provide rigorous evidence for the presence of positive and negative reciprocation among eBay traders. Most importantly, our analysis derives unbiased estimates of the risks that are associated with trading on eBay that, we believe, are more realistic than those suggested by a naïve interpretation of the unusually high (> 99%) levels of positive feedback currently found on that system. 1.

This introduction to the plm package is a slightly modified version of Croissant and Millo (2008), published in the Journal of Statistical Software. Panel data econometrics is obviously one of the main fields in the profession, but most of the models used are difficult to estimate with R. plm is a package for R which intends to make the estimation of linear panel models straightforward. plm provides functions to estimate a wide variety of models and to make (robust) inference. Keywords:˜panel data, covariance matrix estimators, generalized method of moments, R. 1.

An increasing number of reports of online auction fraud are of growing concern to auction operators and participants. In this research, we discuss reserve price shilling, where a bidder shills in order to avoid paying auction house fees, rather than to drive up the price of the final bid. We examine the effect that premium bids, since they are linked with reserve price shill bids, have upon the final selling price. We use 10,260 eBay auctions during April 2001, and identify 1,389 auctions involving 493 sellers and 1,314 involved in concurrent auctions that involving the exact same item. We find that premium bidding occurs 23 % of the time, in 263 of the 1,389 auctions. Using a theoretical perspective involving valuation signals, we show that other bidders may view high bids as signals that an item is worth more. Thus, they may be willing to pay more for the item than items that do not receive premium bids. The implications are disturbing in that sellers may be more motivated to enter a shill bid in order to drive up the final price in an online auction. We also examine and report on alternative hypotheses involving winner’s curse and the possibility of reserve price shill bids. Our results are developed in the context of a weighted least

Trade theories covering Preferential Trade Agreements (PTAs) are as diverse as the literature in search of their empirical support. To account for the model uncertainty that surrounds the validity of the competing PTA theories, we introduce Bayesian Model Averaging (BMA) to the PTA literature. BMA minimizes the sum of Type I and Type II error, the mean squared error, and generates predictive distributions with optimal predictive performance. Once model uncertainty is addressed as part of the empirical strategy, we report clear evidence of Trade Creation, Trade Diversion, and Open Bloc effects. After controlling for natural trading partner effects, Trade Creation is weaker – except for the EU. To calculate the actual effects of PTAs on trade flows we show that the analysis must be comprehensive: it must control for Trade Creation and Diversion as well as all possible PTAs. Several prominent control variables are also shown to be robustly related to Trade Creation; they relate to factor endowments and economic policy.

In this paper, the effects of socioeconomic characteristics (gender, age, education level, marital status, income, occupational and health status, household size and social relationships) on individuals ´ health status in Spain from 1994 to 2001 are analysed. The estimations are carried out using ordered probit models and new data from the whole waves of the European Community Household Panel (ECHP) have been used. The results indicate that personal characteristics, education level, income as well as health status and social relationships have strong influence on self−assessed health. The authors would like to acknowledge the help given by the Centre for Health Economics (CHE) of the University of York (United Kingdom). Also, we are very grateful for many helpful comments from the participants in the York Seminars in Health

Knowledge of how travelers respond to spikes in gas prices is key to planning for future instabilities in gas prices and offers insight into possible pricing strategies as mechanisms for reducing vehicle travel or improving efficiency of fuel use. A survey of over 500 residents in Austin, Texas capitalized on a severe spike in gas prices that transpired in September of 2005. This work examines how respondents ’ travel behavior changed during and following the spike. The paper describes the findings using basic descriptive statistics and uses ordered probit and binary logit models to determine which factors are responsible for behavioral changes in response to gas price spikes. Respondents indicated a strong tendency to reduce overall driving and/or chain together activities in more efficient tours as a way of coping with high prices, and nearly every gas-saving behavior questioned exhibited a significant percentage of persons reporting an increase. The results suggest that urban form, more than demographics, dictates the behavioral responses adopted by individual respondents. Finally, in the wake of the spike, respondents suggested many reasons for the price shifts and voiced support for policy measures that would encourage more efficient fuel use.

This paper examines the effectiveness of multinational enterprises ’ capital budgeting decisions as compared to the decisions of purely domestic enterprises. This is an important question because of multinationals ’ role in allocating capital globally. Answering this question may also shed light on whether multinationals are indeed better managed than are purely domestic firms. We examine this question empirically using the deviation of a firm’s estimated marginal Tobin’s q from an appropriate benchmark as an indicator of effective resource allocation. We find that multinationals make more efficient capital budgeting decisions than do purely domestic firms. The result stems from multinational enterprises ’ exercising greater restraint on over-investment, but is not due to looser liquidity constraints. In obtaining the result, we account for the impact of institutional ownership, managerial ownership, and managerial entrenchment. We also test whether multinationals ’ greater capital budgeting efficiency might be due to their investment locations, since they might thereby be monitored by more agents and also may be more successful in resisting pressures from special interest groups and governments to adopt practices that are not consistent with firm value maximization. We do not find support for the monitoring and bargaining hypotheses. Our observations therefore suggest that multinationals may be intrinsically better managed firms than are purely domestic firms. (JEL F23, G31) 1

Software selection is an important consideration in risk management for information security. Additionally, the underlying robustness and security of a technology under consideration has become increasingly important in total cost of ownership and other calculations of business value. Open source software is often touted as being robust to many of the problems that seem to plague proprietary software. This study seeks to empirically investigate, from an information security perspective specific security characteristics of open source software compared to those of proprietary software. Software vulnerability data spanning several years are collected and analyzed to determine if significant differences exist in terms of inter-arrival times of published vulnerabilities, median time to release ‘fixes ’ (commonly referred to as patches), type of vulnerability reported and the respective severity of the vulnerabilities. It appears that both open source and proprietary software are each likely to report similar vulnerabilities and that open source software is only marginally quicker in releasing patches for problems identified in their software. The arguments favoring the inherent security of open source software do not appear to hold up to scrutiny. These findings provide evidence to security managers to focus more on holistic software security management, irrespective of the proprietary-nature of the underlying software.

Economic development implies that the efficiency of firms in developing countries is approaching that of firms in advanced economies. We examine the extent of this convergence in the Czech Republic and Russia, economies that represent alternative models of implementing development policies, often referred to as the Washington Consensus, that have promoted privatization, competition and foreign investment. We also test hypotheses positing that only firms near the efficiency frontier benefit from these policies and catch up. Using 1992-2000 panel data on virtually all industrial firms in each country, we find that privatization to domestic owners did not markedly improve the efficiency of firms; domestic firms are not catching up to the (world) efficiency standard given by foreign-owned firms; and the distance of the Russian firms to the efficiency frontier is much larger than that of the Czech firms and continued to grow for most firms beyond 1997 while remaining constant in the Czech Republic. Domestic firms closer to the frontier are not more likely to catch up than firms further from the frontier although foreign firms do exhibit this behavior. Foreign-owned firms are increasingly displacing domestic firms in the top deciles of the overall distribution of efficiency, due in part to slower “learning ” by domestic firms, higher