Aktuelle Themenvorschläge für Abschlussarbeiten im Bereich Ökonometrie
Short description: The task is to investigate vulnerability to climate change and willingness to adapt to changes induced by climate change as determinants of economic growth. For the empirical analysis, students may use data provided by the Notre Dame Global Adaptation Initiative (ND-GAIN).
Relevant literature:
Barro, Robert, (2003), Determinants of Economic Growth in a Panel of Countries, Annals of Economics and Finance, 4, issue 2, p. 231-274.
Dell, Melissa, Jones, Benjamin and Olken, Benjamin, (2012), Temperature Shocks and Economic Growth: Evidence from the Last Half Century, American Economic Journal: Macroeconomics, 4, issue 3, p. 66-95.
Bachelor or Master Thesis
Short description: Besides its effect on the real economy, vulnerability to climate change may also have an impact on financial markets. The scope of this thesis is to examine the effect of climate change vulnerability on FDI flows and foreign bank participation using a large panel of countries.
Relevant literature:
Beirne, John & Renzhi, Nuobu & Volz, Ulrich, 2021. "Feeling the heat: Climate risks and the cost of sovereign borrowing," International Review of Economics & Finance, Elsevier, vol. 76(C), pages 920-936.
Master Thesis
Short description: While there exists a large body of evidence on the effect of financial development on growth volatility, most works do not cover the period of and after the financial crisis 2008-09. The aim of this thesis is to replicate existing research on determinants of growth volatility using up-to-date country-level data.
Relevant literature:
A., Denizer Cevdet, Iyigun, Murat and Owen, Ann, (2002), Finance and Macroeconomic Volatility, The B.E. Journal of Macroeconomics, 2, issue 1, p. 1-32.
Kose, M. A., E. S. Prasad, and M. E. Terrones (2003). Financial integration and macroeconomic volatility. IMF Staffpapers 50(1), 119–142.
Bachelor or Master Thesis
Short description: Die Aufgabe besteht darin, Mittels des „BEKKs“ R-Paketes von (Fülle et.al., 2022) eine zwei-Schritt Schätzung durchzuführen. Dabei soll zunächst ein BEKK(1,1,1) Modell geschätzt werden. Der zweite Schritt der Schätzung benutzt die geschätzten Residuen des BEKK Modells und passt daran eine orthogonalisierte Copula-Verteilung an. Die Schätzung soll für verschiedene Copulas durchgeführt und interpretiert werden werden.
Relevant literature:
Lee, Tae Hwy and Long, Xiangdong, (2009), Copula-based multivariate GARCH model with uncorrelated dependent errors, Journal of Econometrics, 150, issue 2, p. 207-218.
Master Thesis
Short description: Die Aufgabe besteht darin, den Value-at-Risk (VaR) und Expected-Shortfall (ES) für verschiede BEKK(1,1,1) Modelle mittels eines Backtestings zu vergleichen. Dabei sollen verschiedene Portfoliozusammensetzungen aus Bitcoin, Gold und S&P 500 untersucht werden. Für die Schätzung soll das „BEKKs“ R-Paketes von (Fülle et.al., 2022) benutzt werden. Für das Backtesting sind die Arbeiten von (Christoffersen, 1998) und (Du and Escanciano, 2015) als Grundlage vorgesehen.
Relevant literature:
Orhan, M., & Köksal, B. (2012). A comparison of GARCH models for VaR estimation. Expert Systems with Applications, 39(3), 3582-3592.
Christoffersen, P. F. (1998). Evaluating Interval Forecasts. International Economic Review, 39(4), 841–862. https://doi.org/10.2307/2527341
Du, Z., & Escanciano, J. C. (2017). Backtesting expected shortfall: accounting for tail risk. Management Science, 63(4), 940-958.
Master Thesis
Short description: The financial crisis, the sovereign debt crisis, the recent pandemic, and perhaps the upcoming energy and food crisis have led to a renewed interest in the possible relationship between time-varying economic uncertainty, business cycle fluctuations, and monetary policy conduct. An important step to answer this question is to empirically quantify the uncertainty, which is an unobservable second-order statistic by its nature. The objective of the thesis is to exploit a data-rich environment to construct a measure of uncertainty in the Euro area. The student is expected to replicate the study of Mumtaz and Theodoridis (2018) -- based on either the original or a reasonably modified version of their model specification -- for the Euro area..
Relevant literature:
Jurado, K., Ludvigson, S. C., & Ng, S. (2015). Measuring uncertainty. American Economic Review, 105(3), 1177-1216.
Mumtaz, Haroon & Theodoridis, Konstantinos, 2020. "Fiscal policy shocks and stock prices in the United States," European Economic Review, Elsevier, vol. 129(C).
Master Thesis
Short description: Since the 1980s, public investment quota experienced a global decline, which is often attributed to the dominance of social expenditures and too rigid fiscal rules (e.g. Schuknecht, Zemanek 2021). While earlier studies (Välilä, Mehrotra 2006) resorted to single-country analysis with vector error-correction models (VECM), more recent studies (Téllez et al. 2022) consider panel models in first-differences. Using the R-package pvars, your task would be to combine these two approaches in a panel VECM in order to reveal common long-run reasons for the global decline. The package and the data panel on investment and GDP can be provided; the latter yet requires own data research for further variables.
Relevant literature:
Delgado-Téllez, M., Gordo, E., Kataryniuk, I., and Pérez, J. (2022). The decline in public investment: "social dominance" or too-rigid fiscal rules? Applied Economics, 54(10), 1123-1136.
Mehrotra, A., and Välilä, T. (2006). Public investment in Europe: Evolution and determinants in perspective. Fiscal Studies, 27(4), 443-471.
Schuknecht, L., and Zemanek, H. (2021). Public expenditures and the risk of social dominance. Public Choice, 188(1), 95-120.
Level: Bachelor or master thesis
Short description: Abu-Qarn and Abu-Bader (2009) have shown how arms race dynamics in the Israeli-Arab conflict can be modeled by vector auto-regression (VAR). Your task would be to (1) replicate their results based on the SIPRI data set and then (2) extend the model by data-driven identification of the structural shocks using the svars package. Own proposals for the analysis of alternative conflicts are welcomed.
Relevant literature:
Abu-Qarn, A. S., and Abu-Bader, S. (2009). On the dynamics of the Israeli–Arab arms race. The Quarterly Review of Economics and Finance, 49(3), 931-943.
Herwartz, H, Lange, A and S. Maxand, S, (2021). Data-driven identification in SVARs - When and how can statistical characteristics be used to unravel causal relationships?, Economic Inquiry, 60(2), 1-26.
Lange, A., Dalheimer, B., Herwartz, H., & Maxand, S. (2021). svars: An R Package for Data-Driven Identification in Multivariate Time Series Analysis. Journal of Statistical Software, 97(5), 1-34.
Level: Bachelor or master thesis
Short description: Your task would be to (1) model the interaction between "economic growth" and the index "quality of liberal democracies" using a bivariate VAR model and then (2) identify economic shocks using Cholesky decomposition or independent component analysis (ICA). According to the hypothesis to be tested, economic crises may have a negative impact on the quality of democracies, just as the mediterranean EU-countries after 2010 let presume. The index data has been very recently assembled by the V-Dem Project, which is a popular data base in political science.
Relevant literature:
Morlino, L., & Quaranta, M. (2016). What is the impact of the economic crisis on democracy? Evidence from Europe. International Political Science Review, 37(5), 618-633.
Level: Bachelor or master thesis
Short description: High-frequency changes in financial assets identified within a tight window around monetary policy announcements have been shown to carry relevant information for monetary policy shocks. While they are often treated as market-based shock measures and used as instruments for identification purposes, a growing body of literature raises concerns about the multi-dimensionality regarding the informational content of these instruments (see e.g. Nakamura and Steinsson, 2018, QJE; Miranda-Agrippino and Ricco, 2021, AEJ:Macro; Jarocinski and Karadi, 2020, AEJ:Macro; Cieslak and Schrimpf, 2019, JIE). The objective of the thesis is to uncover the structural factors embedded in policy announcements of the European Central Bank and exploit their macroeconomic implications. The student is expected to partially replicate the result in Altavilla et al. (2019, JME).
Relevant literature:
Nakamura, E., & Steinsson, J. (2018). High-frequency identification of monetary non-neutrality: the information effect. The Quarterly Journal of Economics, 133(3), 1283-1330.
Miranda-Agrippino, S., & Ricco, G. (2021). The transmission of monetary policy shocks. American Economic Journal: Macroeconomics, 13(3), 74-107.
Jarociński, M., & Karadi, P. (2020). Deconstructing monetary policy surprises—the role of information shocks. American Economic Journal: Macroeconomics, 12(2), 1-43.
Cieslak, Anna and Schrimpf, Andreas, (2019), Non-monetary news in central bank communication, Journal of International Economics, 118, issue C, p. 293-315.
Altavilla, C., Brugnolini, L., Gürkaynak, R. S., Motto, R., & Ragusa, G. (2019). Measuring euro area monetary policy. Journal of Monetary Economics, 108, 162-179.
Swanson, Eric T., 2021. "Measuring the effects of federal reserve forward guidance and asset purchases on financial markets," Journal of Monetary Economics, Elsevier, vol. 118(C), pages 32-53.
Bachelor Thesis
Short description: Reporting errors occur if the eye-catcher that depicts the level of statistical significance is inconsistent with the reported statistical values. These errors may affect the statement of an article. The scope of this thesis is to examine the articles of the top 50 economic journals with the highest share of reporting errors among their statistical tests manually and give a first indication about the severity of reporting errors. Conclusions if reporting errors occur more in a single variable or for a single model can be drawn. Data is already available and will be provided.
Relevant literature:
Pütz, Peter and Stephan B. Bruns (2021). The (non-)significance of reporting errors in economics: Evidence from three top journals. Journal of Economic Surveys Volume 35 Issue 1.
Bachelor or Master Thesis
Short description: The usage of Bayesian methods in economics is on the rise. Using special keyword searches the aim is to detect trends in time. Additionally, these methods can be analysed for statistical accuracy in comparison to the statistical accuracy of frequentist approaches (e. g. reporting errors). Data is already available and will be provided.
Relevant literature:
-
Currie, Janet et al. (2020)
. Technology and Big Data Are Changing Economics: Mining Text to Track Methods. AEA Papers and Proceedings Volume 110.
Master Thesis
Short description: Many journals require the authors to publish their data and code. In a first step, one has to develop a method for checking if the authors shared their data and code for real. To meet this end either regular expressions or web scraping might be useful. In a second step the implications for publication bias have to be analysed by drawing counterfactual p-distributions. Assuming that there is less p-hacking among authors who publish their data and code this allows to separate publication bias and p-hacking at least rudimentarily. Data is already available and will be provided.
Relevant literature:
-
Currie, Janet et al. (2020). Technology and Big Data Are Changing Economics: Mining Text to Track Methods. AEA Papers and Proceedings Volume 110.
Bachelor or Master Thesis
Short description: Reporting errors occur if the eye-catcher that depicts the level of statistical significance is inconsistent with the reported statistical values. Their occurrence might indicate a lack of diligence. This thesis aims to check to what extent reporting errors occur e. g. because of transposed digits or in articles where the authors use only a few decimal places or just very few words in the table notes. Data is already available and will be provided.
Relevant literature:
Pütz, Peter and Stephan B. Bruns (2021). The (non-)significance of reporting errors in economics: Evidence from three top journals. Journal of Economic Surveys Volume 35 Issue 1.
Bachelor or Master Thesis
Short description: Regular expression and other text-mining techniques allow us to examine trends in the research landscape in economics, e. g. the usage of pre-analysis plans or the increase of observations. To meet this end a newly developed tool called DORIS can be used and also be extended. Data is already available and will be provided.
Relevant literature:
-
Currie, Janet et al. (2020). Technology and Big Data Are Changing Economics: Mining Text to Track Methods. AEA Papers and Proceedings Volume 110.
Master Thesis
Short description: There is increasing evidence that at least some of the published results in economics are subject to selective reporting. By using Caliper tests on the published t-values, indications of p-hacking can be unravelled. Additionally, these tests shall be embedded in a regression framework in order to analyse the effect of potential covariates. Lastly, a difference-in-difference design shall give hints about the impact of open data and code policies by the journals on selective reporting. Data is already available and will be provided.
Relevant literature:
-
Bruns, Stephan B. et al. (2019). Reporting errors and biases in published empirical findings: Evidence from innovation research. Research Policy Volume 48 Issue 9.
Master Thesis
Short description: Germany is lacking official statistics for regional prices. The Federal Statistical Office of Germany recently obtained scanner data from the most important retailers that can be used for calculating regional price indices. In a next step the most important drivers for regional price imparities are determined. Data is already available and will be provided.
Relevant literature:
-
Happ, Alina and Chris-Gabriel Islam (2021). The use of scanner data to compile regional price indices in Germany with an outlook to income inequality. In: Proceedings 63rd ISI World Statistics Congress, 11-16 July 2021, Virtual.
Bachelor or Master Thesis
Short description: The price change of seasonal goods, e. g. pumpkins or melons, is difficult to measure due to the lack of availability in certain months. The Federal Statistical Office of Germany recently obtained scanner data from the most important retailers that offer new possibilities to measure the price change of seasonal goods. In a first step, literature about the possible calculation methods shall be gathered. In a second step, these methods shall be used to calculate price indices for seasonal goods. Data is already available and will be provided.
Relevant literature:
-
Diewert, W. Erwin et al. (2022). Seasonal Products. Discussion Paper 22-01, Vancouver School of Economics.
Radjabov, Botir and Ken Van Loon (2022). Seasonal Products and Multilateral Methods. 17th Meeting of the Ottawa Group, 07-10 June 2022.
Bachelor or Master Thesis
Short description: The Federal Statistical Office of Germany recently obtained scanner data from the most important retailers that offer new possibilities to measure consumer prices. Although there are some decent price index calculation methods, it is difficult to measure the quality of each method and to compare them. In a first step, literature about the possible calculation methods shall be gathered. In a second step, these methods shall be used to calculate price indices. In a third step, a measurement for the quality of these indices shall be implemented and applied. Data is already available and will be provided.
Relevant literature:
Master Thesis
Short description: In the measurement of the consumer price index in Germany each article that is observed has to be classified according to the COICOP. The research question is to check the severity of misclassifications related to the overall consumer price index. This has to be checked practically for the current, traditional method following a Laspeyres index but also theoretically for the future methods based on scanner data using multilateral methods, e. g. the GEKS. Data is already available and will be provided.
Relevant literature:
Master Thesis
Short description: The consumer price index is one of the most important economic indicators in Germany which is produced in a joint work between the Federal Statistical Office of Germany and the State Statistical Offices. In order to guarantee a high quality there are certain requirements that have to be met, e. g. the variety of goods or the variety of regional dispersion. In this work, the task is to take a closer look at the calculation and check if the requirements are met or if there is potential for improvement. Data is already available and will be provided.
Relevant literature:
-
Sandhop, Karsten and Timm Behrmann (2016). Weiterentwicklung der Stichprobe in der Verbraucherpreisstatistik. WISTA Wirtschaft und Statistik 5/2016.
Bachelor or Master Thesis
Short description: Price data collected by the Federal Statistical Office of Germany has to be classified to the COICOP. Currently, this is done using machine learning. As the COICOP is hierarchical there is room for improvements of the algorithms. This thesis aims to increase the accuracy and the macro F1 score of the classification. Data is already available and will be provided.
Relevant literature:
-
Silla, Carlos N. and Alex A. Freitas (2011). A survey of hierarchical classification across different application domains. Data Mining and Knowledge Discovery volume 22, pages 31–72.
Master Thesis
Short description: Price data collected by the Federal Statistical Office of Germany has to be classified to the COICOP. Currently, this is done using supervised machine learning. There might be room for improvements of the algorithms using unsupervised learning. This thesis aims to increase the accuracy and the macro F1 score of the classification. Data is already available and will be provided.
Relevant literature:
-
Hastie, Trevor et al. (2009). The Elements of Statistical Learning. Data Mining, Inference, and Prediction, Second Edition. Springer Series in Statistics.
Master Thesis
Short description: Price data collected by the Federal Statistical Office of Germany has to be classified to the COICOP. Currently, this is done using supervised machine learning. Nonetheless, the data is imbalanced in the sense that for some COICOP classes only a few observations can be used for training the algorithms. This thesis aims to increase the accuracy and the macro F1 score of the classification by expanding the training data synthetically. Data is already available and will be provided.
Relevant literature:
-
Chen, Chao et al. (2004). Using Random Forest to Learn Imbalanced Data. Berkeley Statistics Report Number 666.
Master Thesis
Short description: Price data collected by the Federal Statistical Office of Germany has to be classified to the COICOP. Currently, this is done using supervised machine learning. However, the input data is limited due to the information provided by the retailers. This thesis aims to increase the accuracy and the macro F1 score of the classification by generating new input data. This can be done using text-mining techniques onto the article description or by merging metadata that are obtained via scraping or additional data bases. Initial data is already available and will be provided.
Relevant literature:
-
Hastie, Trevor et al. (2009). The Elements of Statistical Learning. Data Mining, Inference, and Prediction, Second Edition. Springer Series in Statistics.
Master Thesis
Short description: Price data collected by the Federal Statistical Office of Germany has to be classified to the COICOP. Currently, this is done using supervised machine learning. However, there might be room for improvement using a principle component analysis (PCA) instead. This thesis aims to increase the accuracy and the macro F1 score of the classification. Data is already available and will be provided.
Relevant literature:
-
Hastie, Trevor et al. (2009). The Elements of Statistical Learning. Data Mining, Inference, and Prediction, Second Edition. Springer Series in Statistics.
Bachelor or Master Thesis