DataPalooza 2022

Results & Supporting Data

Model Name RMSE LM Significant Only 0.14352 Regression Tree 0.11779 Tree no sentiment 0.12164 Neural Network 0.12630 Nnet no sentiment 0.12722 Root Mean Squared Error (RMSE) – quadratic scoring rule which measures the average magnitude of error *Penalizes large errors*

Supporting Literature • Identifying Risk in the Banking System with Sentiment • The Tone of the Beige Book and the Pre ‐ FOMC Announcement Drift • BERT: Pre ‐ training of Deep Bidirectional Transformers for Language Understanding • Sentiment Analysis using deep learning Architectures Results • Tested various forecasting models and found Regression tree yielded the best predictive power (with sentiment scores)

How are we using it now?

1 Pull

2 Analysis 3 Present