UC SAN DIEGO / 32.8801° N, 117.2340° WLOCAL TIME
← WORK INDEX

PROJECT / 004

MarketRegime

Financial Market Regime Detection

DOMAINFinance / Statistical ML
FOCUSfinancial time series · unsupervised learning · evaluation
TOOLSPython · Pandas · NumPy · scikit-learn · statsmodels
ILLUSTRATIVE / SYNTHETIC

01 / Overview

A study of how statistical representations can describe shifts in market behavior over time.

02 / Problem

Regimes should be interpretable and stable enough to be evaluated rather than merely visually compelling clusters.

03 / Approach

Candidate unsupervised and statistical approaches are compared through their assumptions, stability, and explanatory value.

04 / Technical stack

Python · Pandas · NumPy · scikit-learn · statsmodels

05 / Visualization

Illustrative visualization only. It communicates the project direction rather than a completed experimental result.

06 / Future development

Planned checks include robustness across windows and out-of-sample behavior without making predictive claims prematurely.

GITHUB