Zurich Trading Simulator (ZTS)
Background
The Zurich Trading Simulator (ZTS) is an open-source stock exchange simulator that measures dynamic decision-making in financial markets. Developed in 2022 by the Chair of Cognitive Science at ETH Zurich, Sandra Andraszewicz and colleagues, it tracks specific metrics like trading activity (volume, frequency) and risk-taking (proportion of risky assets in a portfolio). By immersing participants in day-to-day trading simulations that mimic real-world financial fluctuations, ZTS can capture both quantitative trading data and qualitative psychological states. Researchers use the platform to explore human reactions to market information, news events, and social influences.
The simulation game consists of a variable round of trading 'years' that each present the participant with the fluctuating daily values of one particular market index for about 252 days that last about 800ms each. The user interface is divided into four sections:
- Market Section: presents a line graph with the daily changing stock value on the y-axis and the trading days on the x-axis
- Portfolio: presents the participant's portfolio information such as amount of cash (safe assets), number of shares (risky assets) and total asset value
- Trading: presents the 'buy' and 'sell' buttons that participants can use to purchase stocks or sell them
- News: can present stock relevant news for the current day
The Millisecond Trading Simulator is based on the ZTS. By default, it plays two rounds of the game using pre-determined pricing paths provided by Andraszewicz and colleagues. While the Millisecond Trading Simulator strives to emulate the ZTS, differences in the implementations exist.
Task Procedure
During a demo/practice round the Millisecond Trading Simulator explains the game, using the ZTS original instructions, and requires participants to make a certain number of purchases and share sells.
The demo/practice phases uses a price path provided by the ZTS that uses a straight-line pattern to ensure that participants stay unbiased in regard to the test stimuli. If participants don't make the minimum requested trades, the demo/practice round is repeated by the Millisecond Trading Simulator. During practice, participants have 1000ms per day to read the News section (where relevant instructions are presented) and make their trades at the daily price value. The demo/practice phase lasts for 118 days.
After the demo/practice phase, the test begins. By default, the Millisecond trading simulator runs two blocks that randomly choose from two price paths of historical market data provided by Andraszewicz and colleagues. These price paths can easily be updated to run different scenarios. Each round runs for 252 days with days 1-252 being actual trading days for the participant (day 0 sets the first share value and assigns chash and shares to the participant). By default, the total assets assigned to the participant on day 0 are $10,000 divided into a cash amount of $5000 (the 'safe' asset) and the maximum number of shares (the 'risky asset) that can be purchased with $5000 on day 0. Each trading day runs for 800ms (default). During that time participants can decide to buy shares (via three buy buttons) or sell shares (via three sell buttons). The daily share price is always prominently displayed in the center of the trading buttons. The buy/sell buttons are worth 1,10, or 20 shares, and participants could potentially request several trades per day if fast enough. The actual number of shares bought/sold depend on the assets that participants own (always displayed in the portfolio section). For example, if a participant requests to buy 20 shares but has only enough cash to cover 18 shares on that particular day, 18 shares are bought. For each requested trade transaction participants receive success or failure feedback for a maximum duration of 300ms. The Millisecond Trading Simulator currently displays 'nothing to report' under the daily news section. Specific daily news can be added manually to the price path data file.
What it Measures
The Millisecond Trading Simulation measures how individuals handle dynamic and risky decision-making in financial market situations
Psychological domains
- Decision-making: Response to potential rewards and losses over time
- Risk-taking: Preference for high-reward/high risk or low-reward/low-risk
- Impulsivity: Ability to self-regulate during market crashes and keep to logical, long-term plans in favor of knee-jerk panic selling or buying
Main Performance Metrics
- Total Asset Value: the cumulative value of safe and risky assets at the end of the round; ultimate metric of a trader’s performance and overall success
- Cash: the amount of the 'safe' asset at the end of the round
- PandL: 'profit and loss' at the end of the round
- ROI: 'return of investment at the end of the round
- Proportion Risky Assets: proportion risky asset (shareValues) relative to the total at the end of the round
- Cash-Equity Ratio: the safe asset relative to the risky Asset at the end of the round
- Trade Counts: number of times participant made a trade (regardless of quantities moved)
- Trade Volume: sum of all transactions made
- Trade Value: number of money moved through the account via buying and selling shared
- Trading response time: mean response time; measure of processing time
Psychiatric Conditions
The ZTS was developed primarily for behavioral economics and experimental finance, its published research relies entirely on non-clinical cohorts.
The Zurich Trading Simulator (ZTS) is an open-source stock exchange simulator that measures dynamic decision-making in financial markets
References
Search Google Scholar for peer-reviewed, published research using the Inquisit Zurich Trading Simulator (ZTS).
Andraszewicz, S., Friedman, J., Kaszás, D., & Hölscher, C. (2023). Zurich Trading Simulator (ZTS) — A dynamic trading experimental tool for oTree. Journal of Behavioral and Experimental Finance, 37, Article 100762. https://doi.org/10.1016/j.jbef.2022.100762
Wichary, Szymon and Allenbach, Monika and Helversen, Bettina von and Sterna, Radosław and Hölscher, Christoph and Kaszas, Daniel and Andraszewicz, Sandra, Skin conductance, risk taking and earnings in a market bubble-and-crash scenario. Available at SSRN: https://ssrn.com/abstract=6882399 or http://dx.doi.org/10.2139/ssrn.6882399
Andraszewicz, S., Kaszás, D., Zeisberger, S. et al. The influence of upward social comparison on retail trading behaviour. Sci Rep 13, 22713 (2023). https://doi.org/10.1038/s41598-023-49648-3