Blog
Developments in automated analysis, new software, research commentary and notes on where the field is going.
How to pivot a table with duplicate key pairs in Pandas (with Code)
Use pivot_table with an explicit aggregation function when repeated row-key and column-key pairs represent legitimate repeated measurements.
How to reshape wide data to long format in Pandas (with Code)
Use melt with explicit identifier and measurement columns to turn each selected wide cell into one long-format row.
How to validate a many-to-one merge in Pandas (with Code)
Use validate="many_to_one" in a left merge to require unique join keys in the right-hand lookup table.
How to aggregate columns with named groupby outputs in Pandas (with Code)
Pass output_name=(input_column, aggregation) pairs to DataFrameGroupBy.agg to create explicitly named summary columns.
The 10 Best Universities for Data Science and Analytics in the US — 2026 Rankings by Career Outcome
Top 10 US data science programmes by career outcomes: MIT Sloan, CMU ML, UC Berkeley MIMS, UChicago Analytics, Columbia DS. Best ROI: Georgia Tech OMSA (~$10.7K total), UIUC MSc DS on Coursera (~$21K). Best for ML research: CMU, MIT, Stanford, Berkeley. Employer relationships and career services drive outcomes more than rankings.
How to flag missing values before imputation in Pandas (with Code)
Capture an isna mask before filling values, then store it as a Boolean or zero-one indicator beside the original column.
How to fill missing values within groups in Pandas (with Code)
Use groupby with transform("median") to return each group median to its original rows, then fill only missing numeric values with that aligned Series.
How to run Tukey pairwise comparisons after ANOVA in Statsmodels (with Code)
Use pairwise_tukeyhsd with the numeric outcomes, their group labels and use_var="equal" to compare all pairs of means with Tukey-adjusted inference.
The Best Free Amazon Product Research Tools for Budget-Conscious Sellers in 2026
Best free Amazon product research stack: Helium 10 Starter (free tier) + Keepa Chrome extension + Amazon Brand Analytics (free for brand-registered sellers) + Amazon autocomplete for keyword research. Free tools have 30–50% revenue estimate error — use them for initial screening, invest in paid tools only for shortlisted products.
How to calculate power for an independent-samples t-test in Statsmodels (with Code)
Use TTestIndPower.solve_power with a planned standardized mean difference, significance level, target power and group allocation ratio to solve for the first group size.