Zur Zeit wird gefiltert nach: Economics
Filter zurücksetzen
Demand Models with Price Endogeneity and Advertising
Christoph Nagel
Center for Doctoral Studies in Economics (CDSE), Fakultät für VWL, Lehrstuhl für Statistik
Universität Mannheim
Für dieses Projekt wurde ein neues tagesaktuelles nationales Haushaltspanel für Deutschland mit 15000 Teilnehmern mit einer Laufzeit von 2 Jahren verwendet. Die Waschmittelkäufe der Haushalte und ihre exakten Kontakte mit TV Werbung sind enthalten.
Der Datensatz wird zur Schätzung von zwei unterschiedlichen Nachfragemodellen mit vertikaler Differenzierung verwendet. Die Schätzungen sämtlicher RC Mixed Logit Modelle wäre ohne den Einsatz des BW-Grids nicht möglich gewesen. Der Speicher universitätsüblicher Rechner ist nicht ausreichend und die Schätzung einer Spezifikation über Simulated Maximum Likelihood beschäftigt einen modernen PC (Intel Core 2 Duo) bis zu 10 Tage.
Modell 1 Ergebnis: Die Preisendogenitätskorrektur nach Petrin und Train (2006) lässt wichtige Endogenitätsgründe außer Acht. Insgesamt zeigt sich, dass in Modellen nur mit der Preisendogenitätskorrektur die Preiselastizitäten unterschätzt werden, wenn man das Ergebnis mit der Schätzung unter Berücksichtigung aller Endogenitätsgründe vergleicht. Die quantitativ wichtigsten Endogenitätsgründe sind Zustandsabhängigkeit/Gewohnheiten und die Produktdetails.
Die Verwendung eines nationalen Split Samples mit Hausman Instrumenten ist ein erfolgreicher Ersatz für fehlende Großhandelspreis-Instrumente. TV Werbung und Haushaltslager sind neutral in Beug auf die Korrektur. Dieses Ergebnis ist neu in der Literatur, in der TV Werbung als beliebtes Beispiel für Preisendogenität genannt wird. TV Werbung erhöht die Preiselastizität nur geringfügig. Wenn Produktdetails vollständig modelliert sind, sinken die Preiselastizitäten. Zustandsabhängigkeit/Gewohnheiten verstärken Preiselastizitäten.
Modell 2 Ergebnis: Lokale TV Werbung und lokale Einzelhandelsaktivität sind auf Marktebene in einem Logit Market Modell "robust" signifikant und erklären Unterschiede zwischen den geographischen Märkten. "Robust" meint, dass dabei 120 verschiedene Datensetups mit mehreren Modellen ausgewertet wurden, um diese Ergebnisse erzielen. Dabei ist die Anzahl der Werbekontakte, nicht aber der Werbedruck (Kontakte relativ zur zeitgleichen Gesamtkonkurrenz) das beste Werbemaß.
Für 4 Datensetups wird in einem Random Coefficients Market Modell die Werbesignifikanz bestätigt und Evidenz für eine heterogene Reaktion der Teilmärkte auf Werbung gefunden.
Link
Software
- Stata, Matlab
Publications
- Nagel; Demand Models with Price Endogeneity and Advertising; Disseration, Universität Mannheim, 2010; pdf
Social Networks and the Process of Globalization
Georg Duernecker
Fakultät für VWL
Universität Mannheim
We propose a stylised dynamic model to understand the role of social networks in the phenomenon we call 'globalization'. This term refers to the process by which even agents who are geographically far apart come to interact, thus overcoming what would otherwise be a fast saturation of local opportunities. A key feature of our model is that the social network is the main channel through which agents search and exploit new opportunities. Thus only if the social network becomes global (heuristically, reaches far) can global interaction be steadily sustained. To shed light on the conditions under which such a transformation may, or may not, take place is the main objective of the paper. One of our interesting insights is that in order for a local social network to turn global, the economy needs to display a degree of geographical cohesion that is neither too high (for then global opportunities simply do not arise) nor too low (in which case there is too little social structure for the process to take off). And if globalization does arise, we show that it often occurs abruptly and consolidates as a robust state of affairs. We also show how it is affected by improvements in the flow at which information travels in the network, or the range at which the social network helps to monitor behavior.
Link
Software
- Matlab
Publications
- G. Duernecker and F. Vega-Redondo; Social networks and the process of globalization; 2012, submitted pdf
Liquidity Commonality, Systematic Fund Flows and Optimal Mutual Fund Behavior
Andy Koch
McCombs School of Business
University of Texas at Austin
Stefan Ruenzi
Chair of International Finance
University of Mannheim
Laura Starks
McCombs School of Business
University of Texas at Austin
Florian Weigert
Chair of International Finance
University of Mannheim
In this paper we investigate the influence of liquidity commonality and systematic fund flows on the investment strategy and performance of mutual funds. We hypothesize that in time periods of low (high) absolute fund netflows, funds - holding shares with high liquidity commonality - perform better (worse) than funds - holding shares with low liquidity commonality. If the netflows of a fund are only weakly correlated with netflows of other funds, we propose that an optimal strategy is to invest in shares with a high degree of liquidity commonality.
- Koch, A.; Ruenzi, S.; Starks, L.; Weigert, F. (2011). Liquidity Commonality, Systematic Fund Flows and Optimal Mutual Fund Behavior, Working Paper, UT Austin and University of Mannheim
Area: Finance
Software:
- Matlab, Stata
Links:
Individuelle Präferenzen und Portefeuillewahl
Hans-Martin von Gaudecker
Fakultät für VWL, Lehrstuhl für Mikroökonometrie
Universität Mannheim
Gemeinsam mit Arthur van Soest (Tilburg) und Erik Wengstrom (Lund) untersuchen wir den Erklärungsgehalt aus ökonomischen Laborexperimenten geschätzter Präferenzparameter in Haushaltsportefeuillewahlmodellen. Zudem vergleichen wir die Verteilungen solcher Präferenzparameter und der Qualität von Entscheidungen in klassischen Laborexperimenten mit Studenten und in der allgemeinen Bevölkerung.
Area: Economics
Software:
- Python, Fortran, Stata
Links:
Extreme Dependence Structures and the Cross-Section of Expected Stock Returns
Stefan Ruenzi
Chair of International Finance
University of Mannheim
Florian Weigert
Chair of International Finance
University of Mannheim
Theory suggests that investors holding stocks with high sensitivities to extreme market downside movements demand additional compensation. Standard asset pricing models are unable to capture these extreme dependencies because they rely on the lin- ear correlation as their sole dependence measure. We show that extreme dependence structures in the form of upper and lower tail dependence are as important as linear dependence in explaining the cross-sectional variation of expected stock returns. Con- trolling for market beta, stocks with lower (upper) tail dependence have high (low) average returns. These effects are different from known market anomalies and cannot be explained by size, book-to-market, momentum, coskewness or downside beta.
- Ruenzi, S.; Weigert, F. (2011). Extreme Dependence Structures and the Cross-Section of Expected Stock Returns, Working Paper, University of Mannheim
Area: Finance
Software:
- Matlab, Stata
Links:
Discrete Choice Models Estimation
Isabel Ruhmer
Center for Doctoral Studies in Economics (CDSE)
Universität Mannheim
Pia Dovern-Pinger
Center for Doctoral Studies in Economics (CDSE)
Universität Mannheim
The aim of this project is to explore consumer behavior in restaurants. Using discrete choice models, we analyze how menu choice is influenced by certain menu- and situational-specific characteristcs. In particular, we investigate whether in addition to prices also the menu design - namely the ordering and grouping of the dishes - influences consumer choice.
First results indicate that consumers tend to be extremeness averse. While overall demand decreases in prices, individuals do not pick alternatives at the lowest or highest end of the price spectrum. Furthermore, the overall number of dishes displayed under a certain sub-category - such as fish or vegetarian - influences a category's choice probability.
As the dataset consists of all bills of a German specialities restaurant over a period of more than seven years, it comprises a huge variety of choices. In total, the data contain menu choices of more than 88,000 individuals who chose among more than 76 alternatives grouped within 6 different categories. As a consequence, the estimations are computationally heavy and could not have been performed without the BwGRID Cluster.
- Ruhmer, Dovern-Pinger: "Do consumers (really) care about prices? Lessons from a restaurant's menu" - to be finished
Area: Economics
Fee Setting Intermediaries: On Real Estate Agents, Stock Brokers, and Auction Houses
Andras Niedermayer
Fakultaet fuer VWL, Lehrstuhl fuer Mikrooekonomische Theorie
Universität Mannheim
Mechanisms where intermediaries charge a commission fee and have the sellers set the price are widely used in practice e.g. by real estate agents, stock brokers, art galleries, or auction houses. We model competition between intermediaries in a dynamic random matching model, where in every period a buyer, a seller, and an intermediary are randomly matched. In any period, every intermediary has a temporary monopoly and designs an exchange mechanism that maximizes his own expected profits. Traders' valuations for the indivisible good depend on their option value of future trade. The following results obtain. First, we show that the intermediary can achieve the highest possible profit with a fee setting mechanism. Second, we characterize when these fees are linear. Third, fee setting is an equilibrium outcome in a dynamic market. Fourth, when the rematching probability increases or, equivalently, the period length decreases, the equilibrium fees become smaller. Our model is applicable to stock brokers and auction houses as intermediaries. It can further explain several of the stylized facts observed in real estate brokerage, such as the 6 percent fee, the relation between listing price and time on market, inefficient free entry, higher prices for houses owned by brokers, and home owners who bought during a boom asking higher prices. We also provide various extensions.
- Andras Niedermayer; Seminar at University of New South Wales; "Fee Setting Intermediaries: On Real Estate Agents, Stock Brokers, and Auction Houses"; online
- Andras Niedermayer; Seminar at Northwestern University; "Fee Setting Intermediaries: On Real Estate Agents, Stock Brokers, and Auction Houses"; online
Area: Economics
Software:
- Matlab, Fortran
Links:
Analyse der Präferenzen von EU-Bürgern zur europäischen Sicherheitspolitik
Bernhard Klingen
Abteilung Volkswirtschaftslehre
Universität Mannheim
Mit einer Mehrebenenanalyse wurden die Präferenzen von EU-Bürgern hinsichtlich der Kompetenzverteilung in der europäischen Sicherheitspolitik untersucht. Im Zentrum standen die konkurrierenden Hypothesen, dass die öffentliche Unterstützung für eine EU- bzw. NATO-Zuständigkeit auf eine Nachfrage nach Sicherheit zurückzuführen ist (Riker-Hypothese) oder dass sie maßgeblich von ideologischen Faktoren (Europhilie, Antiamerikanismus etc.) getragen wird. Als Datengrundlage dienten insbesondere die Umfrageergebnisse der Eurobarometer-Studien EB 62.0 und EB 66.1.
- Klingen, Bernhard; "Is Riker Right? -- Explaining Preferences for Common Defence"; SSRN-Working, 12. August 2009; online.
- Klingen, Bernhard; "A Public Choice Perspective on Defense and Alliance Policy"; in Coyne, Christopher J./Mathers, Rachel L. [Hrsg.], The Handbook on the Political Economy of War; Cheltenham: Edward Elgar Press; 2011; Kap. 17, S. 335-55.
- Klingen, Bernhard; Die Politische Ökonomie der Verteidigungs- und Bündnispolitik; Baden-Baden: Nomos-Verlag; 2011.
Area: Politische Ökonomie
Software:
- Stata
A g-and-h Copula Approach to Risk Measurement in Multivariate Financial Models
Markus Huggenberger
Department of Risk Theory, Portfolio Management and Insurance
Universität Mannheim
Timo Klett
Department of Risk Theory, Portfolio Management and Insurance
Universität Mannheim
We propose and backtest a multivariate Value-at-Risk model for financial returns based on Tukey's g-and-h distribution. This distributional assumption is especially useful if (conditional) asymmetries as well as heavy tails have to be considered and fast random sampling is of importance. To illustrate our methodology, we fit copula GARCH models with g-and-h distributed residuals to three European stock indices and provide results of out-of-sample Value-at-Risk backtests. We find that our g-and-h model outperforms models with less flexible residual distributions and attains similar results as a benchmark model based on Hansen's skewed-t distribution.
Area: Quantitative Finance
Links:
Quality upgrading and price heterogeneity: evidence from Brazilian exporters
Lisandra Flach
Public Finance and Economic Policy Group
Center for Doctoral Studies in Economics, CDSE
University of Mannheim
This project uses a Brazilian firm-product level data to investigate how firms set export prices across countries and presents evidence that firms use strategies of product differentiation and market segmentation. With trade liberalization and increasing competition in foreign markets, Brazilian firms adapted the quality of their products to more exigent markets and, specially, more productive firms upgraded quality to the rich North countries. First, I show preliminary evidence of firm-product price variation across countries and discuss the different hypotheses that may drive price variation. The empirical analysis supports the hypothesis that firms segment the market and produce different qualities to different countries. Second, I build a model with heterogeneous firms and quality differentiation, in which more productive firms produce higher quality and enter the markets in North and South, while the less productive ones export only to South. With trade liberalization, more productive firms adapt product quality to the North to be able to export and/or to increase exports to the rich North consumers, what is reflected in higher price to the North and higher market segmentation. Third, I test the North/South model using exports to the European Union (North) and to Mercosur (South) over years and a difference-in-difference-in-differences (DDD) strategy, and show that the hypothesis of quality upgrading and market segmentation is confirmed for differentiated goods, but not for homogeneous goods: for differentiated goods, more productive firms segment the market and increase product quality to be able to compete in tougher markets.
The project uses three different datasets of Brazilian firms: (1) the SECEX export data, which contains information on export values and quantities of all Brazilian exporters, by firm, product and destination market; (2) the RAIS employer-employee matched data, with information on worker's characteristics; (3) and the data on foreign ownership from the Brazil's Central Bank.
- Flach, Lisandra. Quality upgrading and price heterogeneity: evidence from Brazilian exporters. Presentations: University of Hohenheim (Dec 2010); University of Tübingen (Jan 2011). Scheduled presentations: University of Tilburg (Feb 2011); University of Mannheim (March 2011); University of Nottingham (April 2011).
Area: Economics
Links:
Kategorien
- [-]Fields (122)
- Astrophysics (5)
- Biology (24)
- Biomechanics (1)
- Biophysics (2)
- Chemistry (36)
- Computer Science (7)
- Economics (11)
- Fluid Dynamics (1)
- Geologic Science (2)
- Mathematics (2)
- Medicine (1)
- Physics (26)
- Political Science (2)
- Social Science (2)
- [-]Universities (110)






