Seminar Forecasting and Analytics in Finance


The seminar is open to Master students.

To successfully pass the seminar you need to write a paper and give a presentation. Papers can be written in either German or English and should have a length of 15-20 (team of two) or 20-25 pages (team of three). For hints on how to write a paper see our guidelines. You need to hand in a printed version and also a digital one (PDF). The seminar talks should be given in English.

The two main parts of your paper and presentation will be (i) explanation of the analytical methodology and (ii) replication of analyses from the key references. You should also provide an introduction, a short summary of the literature (which can be part of the introduction), and some concluding remarks. For most topics, you will need to use a software such as R or Matlab.

Please contact your supervisor to discuss the outline of your paper, your empirical part, and any questions that you may have. For organizational questions, please ask Nenad Curcic.

FAQ & Organisational matters

  • Do we get a grade? Yes. Your paper and your presentation will be graded and lead to one grade (equally weighted). Both the paper and presentation have to be passed.
  • What do we have to hand in? An outline of your paper to discuss the content of your paper and your final paper one week before the presentation.
  • Who is responsible? For content-related questions, please contact your supervisor. For organizational questions, please ask Nenad Curcic.

Time Table

  1. 21.01.2017 - 29.01.2017 Submission of your seminar preferences via online platform:
  2. 30.01.2017 First round of seminar matching
  3. 06.02.2017 Second round of seminar matching
  4. 16.02.2017 Topic allocation (comlete the polls on Moodle until 14.02.)
  5. until 27.02.2017 Registration at the Higher Services Portal
  6. until 27.04.2017 Meet your supervisor to discuss the outline of the paper
  7. 02.06.2017 Submission of the paper until 11:59 am, HeHo 18, room 1.00 (Curcic)
  8. 09.06.2017 - 10.06.2017 Presentations, in Villa Eberhardt (exact schedule will be known on 07.06.2017)



1. Partial Least Squares

Introduce the partial least squares regression technique. Then replicate and update the analysis from the key reference. You do not need to discuss or implement:  the wild bootstrap , the Stambaugh correction,  macroeconomic predictors that are not publicly available, Table 3, Tables 8-11.

Key reference:  

Huang, D., Jiang, F., Tu, J., & Zhou, G. (2014). Investor sentiment aligned: a powerful predictor of stock returns. Review of Financial Studies, forthcoming.

supervised by: Prof. Dr. Gunter Löffler

students: Odunayo Rotimi, Julia Steinmetz and Jolivette Tematio Nguekeu


2. Lasso

Introduce the adaptive Lasso regression technique. Then replicate and update the analysis from the key reference. You do not need to discuss or implement: network analysis (section 2.4) and alternative approaches (section 4).

Key reference:

Rapach, D., Strauss, J., Tu, J., & Zhou, G. (2015). Industry interdependencies and cross-industry return predictability. Available at SSRN 2566541.

supervised by: Prof. Dr. Gunter Löffler

students: Tobias Furtwängler, Alexander Kreienbring and Madeline Marquardt


3. Two-Layer Bias Decision Tree and Trading Rule

You should review the application of decision trees in finance. As a second task, you should replicate and update the studies of Wang and Chan (2006) using daily stock prices of Microsoft and Apple.

Key reference:

Wang, J.-L., & Chan, S.-H. (2006). Stock market trading rule discovery using two-layer bias decision tree. Expert Systems with Applications, 30, 605-611.

supervised by: Nenad Curcic

students: Yannick Wahler, Betina Lorena Vietor and Oyelekan Olorunkosebi


4. Predicting Stock Market Movements with Random Forests

Random forests are a machine learning technique based on decision trees. Kumar and Thenmozhi (2006) apply random forests to the Indian stock market and find a hit ratio of 67.4% when trying to forecast the sign of the S&P CNX NIFTY Index price movement.
You shall apply random forests based on the predictors used by Kumar and Thenmozhi to DAX price movements. It is recommended to use the R package randomForest.

Key references:

Kumar, M., & Thenmozhi, M. (2006, January). Forecasting stock index movement: A comparison of support vector machines and random forest. In Indian Institute of Capital Markets 9th Capital Markets Conference Paper.

Liaw, A., & Wiener, M. (2002). Classification and regression by random forest. R news, 2(3), 18-22.

supervised by: Carsten Schäfer-Siebert

students: Elena Droval, Ilia Khomenko and Andrii Yaminskyi


5. New Approach for Stock Return Estimation

You should apply the new method of Hou, van Dijk and Zhang to estimate the earnings of German firms and then to estimate stock returns using these earnings and analyst’s earnings. For calculating stock returns, you should use the method of Easton. Discuss and compare your results with the relevant literature.

Key references:

Hou, K., van Dijk, M. A., & Zhang, Y. (2012). The implied cost of capital: A new approach. Journal of Accounting and Economics, 53(3), 504-526.

Easton, P. (2004). PE ratios, PEG ratios and estimating the implied expected rate of return on equity capital. The Accounting Review, 79, 73-95.

supervised by: Nenad Curcic

students: Artem Chekin, Tim Nägele and Lucas Teichmann


6. Recession Forecast with Naive Bayes

Davig & Hall present the Naïve Bayes model as a new forecasting method of a recession. Compare this model with other approaches of recession forecast. Additionally, replicate a part of the paper of Davig & Hall taking into account the most important four macro-economic variables.

Key reference:

Davig, T., & Smalter Hall, A. (2016). Recession forecasting using Bayesian classification. Available at SSRN 2821968.

supervised by: Clara Franke

students: Syed Hussain, Anna Kehl and Varduhi Khachatryan


7. The Power of Bagging

In times of Big Data, model uncertainty and questionable data quality demand for new approaches in order to produce reliable results. The OpCaR methodology used for the calibration of the new standardized measurement approach for operational risk makes use of the “bagging”: Value-at-Risk estimates based on different distributional assumptions and fitting approaches lead to an average Value-at-Risk which is assumed to be a more robust result.
You shall implement the OpCaR methodology (e.g. in R) and perform a simulation study. Are the averaged estimates from OpCaR indeed more reliable than a benchmark?

Key reference:

The Basel Committee on Banking Supervision (2014): Operational risk – Revisions to the simpler approaches (

supervised by: Carsten Schäfer-Siebert

students: Elena Kireeva, Wanqi Song and Yucheng Xu


8. The Google Trends Strategy

How can Google Trends be useful for predicting index returns? Determine the cumulative returns of the ‘Google Trends strategy’ by Preis et al. for selected search terms on weekly basis in the years 2011-2016. Further, apply the approach of Preis et al. on the German stock market using the DAX share price index.

Key reference:

Preis, T., Moat, H. S., & Stanley, H. E. (2013). Quantifying trading behavior in financial markets using Google Trends. Scientific reports, 3.

supervised by: Clara Franke

students: Yu Fu, Rui Li and Yingwei Liu


9. Google Correlate and the Gold Price

Scott and Varian (2014) combine a structural time-series model, a spike-and-slab prior and data from Google Correlate for short-term forecasting with application to unemployment rates and retail sales.
You shall apply the approach of Scott and Varian to a different time series of data: Can movements in the gold price be predicted by Google search data?
It is recommended to use the R packages KFAS and runjags.

Key reference:

Scott, S. L., & Varian, H. R. (2014). Predicting the present with bayesian structural time series. International Journal of Mathematical Modelling and Numerical Optimisation, 5(1-2), 4-23.

supervised by: Carsten Schäfer-Siebert

students: Yuxuan Cao, Moritz Krystjanczuk and Pengcheng Wan


10. Social Media as Leading Indicator of Firm Equity Value

In this seminar, you should answer the question: “Is there a significant predictive relationship between social media and firm equity value?” by summarizing results of relevant literature. In the second part, you should replicate and update the section 5.3 in the studies of Luo, Zhang and Duan considering the firms of your choice.

Key reference:

Luo, X., Zhang, J., & Duan, X. (2013). Social Media and Firm Equity Value. Information System Research, 146-163.

supervised by: Nenad Curcic

students: Pascal Frick, Arina Khataniuk and Mayya Rozovskaya 


Registration for this seminar is not possible anymore. 


Gunter Löffler

Nenad Curcic

Carsten Schäfer-Siebert

Clara Franke

Dates and Room

Please note the detailed timetable.

Module description

This seminar is open for Master students.

Module description