Auditing Algorithmic Bias in Transformer-Based Trading
This work addresses bias risks in financial AI applications, but it is incremental as it focuses on auditing an existing model without introducing a new method.
The paper tackled the problem of algorithmic bias in transformer-based trading by auditing the model's reliance on volatile data and quantifying how price movement frequency affects prediction confidence, revealing that the model disregards data volatility and is biased toward lower-frequency price movements.
Transformer models have become increasingly popular in financial applications, yet their potential risk making and biases remain under-explored. The purpose of this work is to audit the reliance of the model on volatile data for decision-making, and quantify how the frequency of price movements affects the model's prediction confidence. We employ a transformer model for prediction, and introduce a metric based on Partial Information Decomposition (PID) to measure the influence of each asset on the model's decision making. Our analysis reveals two key observations: first, the model disregards data volatility entirely, and second, it is biased toward data with lower-frequency price movements.