The seminar will be organized in the form of personal consultations in which the topic selected by the student will be discussed. Student is supposed to work independently and we will discuss his/her progress, future directions of work, literature etc. during the consultations. Student is supposed to propose his/her topic (i.e. the assignment of the problem) he/she wants to solve in the seminar. Some topics are suggested below.
The goal of the seminar is provide step-by-step support for the student so that in the end he or she creates working automated trading system in one of the following platforms: Metatrader 4/5 (see www.metaquotes.net), xStation (see developers.xstore.pro), Amibroker (see www.amibroker.com), Matlab (see www.mathworks.com/products/matlab/) or other satisfactory platform. Knowledge of programming at basic level is required.
Verification of directional changes concept and its application in trading.
See also http://www.bracil.net/finance/papers/Voicu-DCTrading-MSc2012.pdf.
The aim of the seminar is to propose automated trading system based on the technical analysis rules and optimize the parameters in these rules, see e.g. Kozola (2014). The statistical testing of the profitability level (of the optimized automated trading system) is obligatory. Basic reference for this topic is Arson (2007).
The seminar will be focused on back testing of different portfolio optimization models. Based on a selected historical dataset, the composition of the portfolio will be re-optimized for each period applying different models and growth of portfolio value in time will be obtained under the assumption that we rebalance the portfolio after each re-optimization.
The focus of the seminar is the estimation of market risk for a given portfolio. Based on a selected historical dataset and given portfolio composition, the goal of the student is to estimate the risk (i.e. either Value at Risk or Expected Shortfall). The back testing of the risk estimation is obligatory.
The company appraisal by means of discounted cash flow model depends significantly on the estimation of the future cash-flows, which can be predicted based on the prediction of variables such as company market share, gross domestic product, foreign exchange rate etc. However these variables are random, and thus also the result of appraisal should be a random variable. The goal of the seminar is to apply the Monte Carlo simulation in the company appraisal by means of discounted cash flow model.
The elaboration of other topics is possible only after their approval.
ARSON, R. David. (2007). Evidence-Based Technical Analysis: Applying the Scientific Method and Statistical Inference to Trading Signals. New York: Wiley, 2007. 528 s. ISBN 978-0-470-00874-4.