This dissertation consists of four papers in structural empirics that can be broadly categorized into two areas. The first three papers revolve around the structural estimation of demand for differentiated products and several applications thereof (Berry (1994), Berry, Levinsohn and Pakes (1995), Nevo (2000)), while the fourth paper examines the U.S. Treasury yield curve by estimating yields as linear functions of observable state variables (Ang and Piazzesi (2003), Ang et al. (2006)). The central focus of each paper are the underlying economics. Nevertheless, all papers share a common empirical approach. Be it prices of beers in Sweden or yields of U.S. Treasury bonds, it is assumed throughout that the economic variables of interest can be modeled by imposing specific parametric functional forms. The underlying structural parameters are then consistently estimated based on the variation in available data. Consistent estimation naturally hinges on the assumption that the assumed functional forms are correct. Another way of viewing this is that the imposed functions are flexible enough not to impose restrictive patterns on the data that ultimately lead to biased estimates of the structural parameters and thereby produce misleading conclusions regarding the underlying economics. In principle, the danger of misspecification could therefore be avoided by adopting sufficiently flexible functional forms. This, however, typically requires the estimation of a growing number of structural parameters that determine the underlying economic relationships. As an example, we can think of the estimation of differentiated product demand. The key object of interest here is the substitution patterns between the products. That is, we are interested in what happens to the demand of good X and all its rival products, as the price of good X increases. With N products in total, we could collect the product-specific changes in demand in a vector with N entries. It is also possible, however, that the price of any other good Y changes and thereby alters the demands for the remaining varieties. Thus, in total, we are interested in N2 price effects on product-specific demand. With few products, these effects could be estimated directly and the risk of functional misspecification could be excluded (Goolsbee and Petrin (2004)). With 100 products, however, we are required to estimate 10,000 parameters, which rarely, if ever, is feasible. This is the curse of dimensionality. Each estimation method employed in the four papers breaks this curse by imposing functions that depend on relatively few parameters and thereby tries to strike a balance between the necessity to rely on parsimonious structural frameworks and the risk of misspecification. This is a fundamental feature of empirical research in economics that makes it both interesting and challenging. / <p>Diss. Stockholm :  Stockholm School of Economics, 2012. Introduction together with 4 papers</p>

Une question fondamentale qui surgit dans le cadre de l’analyse probabiliste du risque sismique est le choix des indicateurs de nocivité des signaux sismiques. En plus de réduire la variabilité de la réponse structurelle (ou non structurelle),un indicateur amélioré (i.e. capable de mieux capturer les caractéristiques de nocivité des mouvements sismiques, aussi bien que l’alea sismique) fournit des critères moins stricts pour la sélection des signaux sismiques.Deux nouveaux indicateurs sont proposés dans cette étude: le premier, nommé ASAR (i.e. Relative Average Spectral Acceleration), est conçu pour la prévision de la demande structurelle, le second, nommé E-ASAR (i.e.Equipment Relative Average Spectral Acceleration), vise à prévoir la demande des composants non structuraux. Les performances des indicateurs proposés sont comparées avec celles des indicateurs de la littérature, sur la base de: a)milliers d’enregistrements sismiques ; b) analyses numériques conduites avec des modèles représentants différents types de bâtiments; et c) analyses statistiques rigoureuses des résultats. Selon l'étude comparative, les indicateurs développés s'avèrent être plus “efficaces” que les indicateurs couramment utilisés. D'ailleurs, l’ASAR et l’E-ASAR ont montré au propre la caractéristique de la “suffisance” en ce qui concerne la magnitude, la distance source-site, et le type de sol (VS30). De plus, les deux indicateurs originaux peuvent être calculés simplement avec la connaissance de la fréquence fondamentale du bâtiment. Cette caractéristique rend l’ASAR et l’E-ASAR facilement exploitables dans les études probabilistes d’alea sismique.Par conséquent, en raison de leur efficacité, suffisance, robustesse et formulation simple, l’ASAR et l’E-ASAR peuvent être considérés comme des candidats prometteurs pour la définition de l’alea sismique dans les cadres de l'analyse probabiliste et déterministe du risque sismique. / A fundamental issue that arises in the framework of Probabilistic Seismic Risk Analysis is the choice of groundmotion Intensity Measures (IMs). In addition to reducing record-to-record variability, an improved IM (i.e. one able tobetter capture the damaging features of a record, as well as the site hazard) provides criteria for selecting input groundmotions to loosen restrictions.Two new structure-specific IMs are proposed in this study: the first, namely ASAR (i.e. Relative Average SpectralAcceleration), is conceived for Structural demand prediction, the second namely, E-ASAR (i.e. Equipment-RelativeAverage Spectral Acceleration), aims to predict Non-Structural components acceleration demand. The performance ofthe proposed IMs are compared with the ones of current IMs, based on: a) a large dataset of thousands recordedearthquake ground motions; b) numerical analyses conducted with state-of-the-art FE models, representing actualload-bearing walls and frame structures, and validated against experimental tests; and c) systematic statistical analysesof the results. According to the comparative study, the introduced IMs prove to be considerably more “efficient” withrespect to the IMs currently used. Likewise, both ASAR and E-ASAR have shown to own the characteristic of“sufficiency” with respect to magnitude, source-to-site distance and soil-type (Vs30). Furthermore, both the introducedIMs possess the valuable characteristics to need (in order to be computed) merely the knowledge of the building’sfundamental frequency, exactly as it is for the wide-spread spectral acceleration Spa(f1). This key characteristic makesboth ASAR and E-ASAR easily exploitable in Probabilistic Seismic Hazard Analysis.Therefore, due to their proven efficiency, sufficiency, robustness and applicable formulation, both ASAR and EASARcan be considered as worthy candidates for defining seismic hazard within the frameworks of both Probabilisticand Deterministic Seismic Risk Analysis.

[pt] Esse trabalho investiga o papel da descoberta de produto na demanda por jogos eletrônicos. Nós mostramos que o padrão de vendas para jogos eletrônicos é bastante variado, com alguns jogos tendo suas vendas imediatamente depois de seus lançamentos, e outros com caldas de venda maiores. Para entender essas diferenças nós propomos um modelo de demanda em que consumidores são periodicamente informados sobre a existência de um jogo e exploramos o padrão de vendas que isso implica. Em seguida aplicamos o modelo aos dados usando dados de preços e vendas da plataforma digital Steam e dados e busca online do google trends. Nossos resultados sugerem que vendas três meses depois do lançamento são em média metade do que seria caso consumidores tivessem informação completa. / [en] This paper investigates the role of product discovery in the demand for video games. We show that the lifetime sales patterns for video games vary widely, with some games selling most of their units in the first months after launch and others having longer tails. To understand these differences we propose a demand model in which consumers are periodically informed about the existence of a game and explore the lifetime sales patterns that this implies. We then take it to the data using price and sales figures from the Steam digital platform and web search figures from google trends. Our results imply that sales three months after launch are on average half of what they should be were consumers fully informed.