(Serious) Time for Time Series
summary
Uses Pandas to parse timestamps (pd.to_datetime), set the datetime index and resample hourly/monthly/daily, create lag features with shift, and smooth with a rolling mean; builds simple, interpretable forecasts by one‑hot encoding month effects and fitting a linear model inside a scikit-learn pipeline. Demonstrates seasonal/trend/error decomposition on household power and CA air‑quality series, holds out the last two years for test, and shows seasonal dummies (and smoother triangular alternatives) extrapolate into the future with quantifiable residuals.
AI-written summary · about
chapters (37)
- 0:00Start
- 2:35What Is Your Experience with Python
- 3:47What Is Your Job
- 4:58Where Are You from
- 5:29What Are You Hoping To Gain from this Tutorial
- 6:43Introduce Time Series as a Concept
- 8:32Introduction
- 9:35Time Series Analysis with Pandas
- 11:03Time Series Analysis
- 13:07Time Series Utilities and Pandas
- 13:28Timestamps in Pandas
- 15:22Days in Month
- 16:00Convert between Time Zones
- 17:05Convert One Time Zone to another
- 19:19Format the Timestamps
- 24:32Easy Aggregations with Resample
- 29:36Create Time-Based Features
- 31:03Shifting
- 34:05Rolling and Smoothing
- 34:51Rolling Average Smoothing
- 38:25Centering
- 41:10Exponential Smoothing
- 41:59Weighted Smoothing
- 48:13Seasonality Modeling
- 54:43Seasonality
- 55:44Residuals
- 1:02:24Linear Regression
- 1:03:48Fit Linear Regression
- 1:17:53Machine Learning Pipeline
- 1:27:33Average Season Effect
- 1:30:04Error Trend Seasonality Decomposition
- 1:30:35Remove Seasonality
- 1:34:13Summary
- 1:36:29Breakout Rooms
- 1:37:26Example Data Sets
- 1:39:05Closed Captioning
- 1:41:10Bonus Materials
related talks
01h 18m(Serious) Time for Time Series
44mMethods and Tools for Time Series Data Science Problems
01h 30mTime Series Analysis with StatsModels
01h 24mA gentle introduction to Pandas timeseries and Seaborn
26mTime Series Analysis in Python: Easy Tools for Scientific Insight
29mA dive into time series for the energy sector
01h 26mApplying Statistical Modeling & Machine Learning to Perform Time-Series Forecasting
01h 27mIntroduction to Machine Learning with Time Series