The Value of Null Results
summary
Collected a pizza‑delivery dataset of 1301 rows and moves from Excel to Python using Query Storm for SQL queries then pandas and plotting. Tests include a difference‑of‑means t test (p ≈ 0.377) and a linear regression with a distance coefficient ≈ $0.16/mile and R² ≈ 0.01, showing almost no predictive power; a confounder of hidden springs average $5.09 on 33 of 38 far deliveries explains the outliers. The build and analyses succeed at demonstrating useful null results.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 3:10Immigrant roots to analyst
- 8:30Live coding pizza data
- 15:52Statistical tests no predictors
- 23:33Null results in business
- 27:49Q&A
related talks
24mBeyond the P-Value: A Data Scientist’s Guide to Tier-1 Feature Launches
35mIntroduction to data munging with pandas and IPython Notebook
22mHow Deliveroo Use Spatial Data Science To Fuel Growth
25mForecasting critical food violations at restaurants using open data
01h 29mPandas and Plots for Data Analysis
42mUsing Postgres to locate the best coffee near you!
27mUsing embedded SQL engines for plotting massive datasets on a laptop
39mMastering DataFrame Diffing Techniques