DDustin Liang·1d agoGranger Causality in Python with statsmodels: Does News Heat Lead Trading Activity, or Just Look Like It?Two attention series that rise together on a chart invite an obvious question: which one moves first? Granger causality is the standard statistical answer. It asks a narrow, testable thing: do past va12 min read·dustinliang.hashnode.dev
DDustin Liang·1d agoTfidfVectorizer Keyword Extraction in Python: Turning Company Descriptions into Concept TagsKeyword extraction with scikit-learn's TfidfVectorizer looks like a two-liner: fit the vectorizer, sort each row, take the top terms. On a real corpus the first result is usually disappointing. You ge10 min read·dustinliang.hashnode.dev
DDustin Liang·1d agoRolling Spearman Correlation in pandas: Testing Whether Two Stocks' News Attention Moves TogetherRolling Spearman correlation is the rank correlation between two series, recomputed over a trailing window: inside each window both series are converted to ranks, and the Pearson correlation of those10 min read·dustinliang.hashnode.dev
DDustin Liang·1d agoPoint-in-Time Data for Backtesting Theme Strategies: Rebuilding Concept Membership You Can TrustPoint-in-time data for backtesting means every input carries the moment it became knowable, and the backtest only reads rows whose knowable time is earlier than the simulated decision. For theme and c10 min read·dustinliang.hashnode.dev
Aanders lee·3d agoA Texas county's tax rate was doubled in the state file, and two of my three checks passed itThe Texas Comptroller publishes a spreadsheet with the tax rate of all 254 counties and a second one with more than 4,800 local taxing units. I use both to estimate property tax bills, so a wrong rate8 min read·anders.hashnode.dev
YYashraj Verma·5d agoRanking Indian Cities by Website Coverage With pandas (and Why You Should Not Average Percentages)Rank Indian cities by website coverage with pandas groupby rank, weighted shares and nlargest, using real clinic data.3 min read·mantisleads.hashnode.dev
PPranaav Karthik·6d agoUnderstanding ETL in Python Through a Practical ProjectA dashboard is only as good as the data behind it. If dates are text and duplicates are still in the file, every chart is slightly wrong. ETL is the standard way to fix that before any charts exist. I7 min read·pranaavkarthik.hashnode.dev
PPranaav Karthik·6d agoPandas for Beginners: Cleaning and Analysing Real-World DataReal data is messy. Blank cells, repeated rows, column names with spaces in them. If you open a CSV in Excel you can fix these by hand, but that stops working once you have thousands of rows. Pandas i8 min read·pranaavkarthik.hashnode.dev
PPranaav Karthik·6d agoWhat I Learned Studying a Python Sales Analytics Dashboard (Streamlit + Pandas)A student's walkthrough of an open-source Python sales analytics dashboard: ETL, derived columns, KPIs, charts and filters, with credit to its author.7 min read·pranaavkarthik.hashnode.dev
Aanders lee·Nov 9What I learned turning government property tax spreadsheets into one county datasetI wanted a simple answer to a simple question: how much property tax does a typical homeowner pay in each U.S. county, and what are the actual local rates behind it? The data exists, but it's spread a5 min read·anders.hashnode.dev