vickrey-sensitivity

Overview

This repository contains:

• Experiment Scripts (Python): For running human participant experiments and collecting data.
• Data Analysis (MATLAB): For analyzing the collected data.
• Collected Data: Raw data files from N=11 participants.

---

Experiment Scripts

• MAIN.py: Main file to run during a human participant experiment. The program generates randomized trials, robobidder models, a randomized duration, and saves collected data to a .csv file in the same directory. It displays a plot of this collected information upon auction completion (i.e. when randomized time endpoint is fulfilled).
• robo_bidder.py: Contains the class robo_bidder which specifies each robobidder's bidding behavior
• robo_model.py: Contains the class robo_model which specifies a set of robo-bidders parametrized by the # of desired robobidders and their growth rates. These models are generated for the sensitivity analysis.

The .csv file output from MAIN.py records the following columns:

• Time(mins): Time since the start of the Vickrey Auction. Each iterative auction is 2 mins long.
• Subject Bid Amount: Dollar amount the human participant reports to the researcher to input during the experiment.
• Won?: Win status of the participant (TRUE if won, FALSE otherwise).
• Winning Payout Amnt: Dollar amount corresponding to the second lowest bid amongst auction participants (including robo-bidders).
• Total Subject Winnings: Ongoing, tallied sum of the human participant’s winnings throughout the experiment.

---

Reproducibility & Setup

1. Quick Setup (Recommended)

Run the provided setup script to automatically configure the Python environment and view MATLAB setup instructions:

bash setup.sh

This will:

• Create a Python virtual environment (venv) if it doesn't exist
• Install Python dependencies from requirements.txt
• Print MATLAB setup instructions

---

2. Manual Python Environment Setup

If you prefer to set up manually:

Using venv (standard Python)

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

2. MATLAB Environment

• MATLAB Version: R2024b
• Required Toolboxes: Statistics and Machine Learning Toolbox
• All custom MATLAB scripts and functions are included in the MATLAB analysis folder.

3. Data

• Data files are in the Collected Data folder.

---

How to Run

Python Experiment

1. Set up the Python environment as above.
2. Run the experiment:

   python Experiment Scripts/MAIN.py

3. Data will be saved as .csv in the same directory.

MATLAB Analysis

1. Open MATLAB.
2. cd into the Data analysis folder.
3. Run the MAIN.mlx script for data analysis.

---

Versioning

• Python: 3.9
• MATLAB: R2024b
• OS: macOS, Windows

---

Citation

If you use this code or data, please cite this repository and the associated Zenodo DOI.