310 Sammlungen allgemeiner Statistiken
Filtern
Schlagworte
- Analyse (1)
- Baye´sche-Statistik (1)
- Datenerhebung auf Keyword-Ebene (1)
- Erhebung (1)
- Online-Marketing (1)
- Umfrage (1)
- bezahlte Suchkampagne (1)
- keyword-level data (1)
- paid search campaigns (1)
Online marketing, especially Paid Search Advertising, has become one of the most important paid media channels for companies to sell their products and services online. Despite being under intensive examination by a number of researchers for several years, this topic still offers interesting opportunities to contribute to the com- munity, particularly because of its large economic impact and practical relevance as well as the detailed and widely unfiltered view of consumer behavior that such marketing offers. To provide answers to some of the important questions from advertisers in this con- text, I present four papers in my thesis, in which I extend previous works on optimization topics such as click and conversion prediction. I apply and extend methods from other fields of research to specific problems in Paid Search. After a short introduction, I start with a paper in which we illustrate a new method that helps advertisers to predict conversion probabilities in Paid Search using sparse keyword- level data. We address one of the central problems in Paid search advertising, which is optimizing own investments in this channel by placing bids in keyword auctions. In many cases, evaluations and decisions are made with extremely sparse data, al- though anecdotal evidence suggests that online marketing is a typical
This cumulative thesis extends the econometric literature on testing for cointegration in nonstationary panel data with cross-sectional dependence. Its self-contained chapters consist of two publications and two publication manuscripts which present three new panel tests for the cointegrating rank and an empirical study of the exchange rate pass-through to import prices in Europe. The first chapter introduces a new cointegrating rank test for panel data where the dependence is assumed to be driven by unobserved common factors. The common factors are first estimated and subtracted from the observations. Then an existing likelihood-ratio panel cointegration test is applied to the defactored data. The distribution of the test statistic, computed from defactored data, is shown to be asymptotically equivalent to that of a test statistic computed from cross-sectionally independent data. The second chapter proposes a new panel cointegrating rank test based on a multiple testing procedure, which is robust to positive dependence between the individual units’ test statistics. The assumption of a certain type of positive dependence is shown by simulations not to be violated in panels with dependence structures commonly assumed in practice. The new test is applied to find empirical support of the monetary exchange rate model in a panel of eight OECD countries. The third chapter puts forward a new panel cointegration test allowing for both cross-sectional dependence and structural breaks. It employs known individual likelihood-ratio test statistics accounting for breaks in the deterministic trend and combines their p-values by a novel modification of the Inverse Normal method. The average correlation between the probits is inferred from the average cross-sectional correlation between the residuals of the individual VAR models in first differences. The fourth chapter studies the exchange rate pass-through to import prices in a panel of nineteen European countries through the prism of panel cointegration. Empirical evidence supporting a theoretical long-run equilibrium relationship between the model’s variables is found by the newly proposed panel cointegration tests. Two different panel regression models, which take both cointegration and cross-sectional dependence into account, provide most recent estimates of the exchange rate pass-through elasticities.