Statistical Inference of Hidden Markov Models on High-Frequency Quote Data
Model Formulation
- Construct Hidden Markov Models on the top of the book data
- Observation Vector at
$t_i$ - Bid Size
- Offer Size
- OrderBook Imbalance
- Spread
Statistical Inference
- Baum-Welch Algorithm
- Viterbi Dynamic Programming Algorithm
Numerical Results
- For each feature, HMM was fitted on a single day of high-frequency top-of-book data
- Procedure repeated for the entire month of Jan 2020
- Optimal parameter estimates from PSG and Hmmlearn were compared using two sample t-tests at a 5% significance level
Repo Outline
- Data Preprocessing and Feature Generation
- Inference Problem and Helpers
- Summary Report and Presentation