Tracking Policy Evolution Through Clustering
summary
Introduces a reproducible pipeline that converts continuous macro series into categorical trajectories using quartile-based categorization, computes similarity with a modified Hamming distance, and reports a 72% hamming diversification index across North America, Western Europe and Scandinavia. Implements everything in Python (pandas, scikit-learn, matplotlib), applies it to 60+ years of fiscal data for 8 countries, produces volatility/persistence KPIs, and shows the method highlights when policy effects diverge.
AI-written summary · about
chapters (13)
- 0:00Macroeconomic policy tracking
- 2:17Key macroeconomic links
- 4:03Economic crisis lessons
- 7:15Reinhardt-Rogoff debt hypothesis
- 10:50Debt and inflation confounders
- 13:41Dynamic clustering framework
- 16:51Quartile analysis and volatility
- 19:44Short-term policy implications
- 22:12Trajectory similarity metrics
- 24:42Summary and paper access
- 25:03Q&A: Sustainable debt levels
- 27:53Q&A: Beyond economics
- 30:05Q&A: Fed influence
related talks
01h 31mPandas, Data Wrangling & Data Science
28mMall Analytics Using Telco Data & Pandas
01h 43m(Serious) Time for Time Series
32mData Transformation: A Framework for Exploratory Data Analysis
31mUsing Simpson’s Paradox to Discover Interesting Patterns in...
39mTemporal Analysis on Topics Utilizing Word2Vec
02h 23mEffective Data Visualization
25mUsing Markov Chain approximation to identify financial fraud