Rolling market-exposure estimates with statsmodels

Published by statsmodels developers

This credits the original publisher. Better Loop membership or a shared assessment is not implied.

The public work

The notebook applies a 60-month rolling CAPM to technology-industry excess returns using Ken French’s factor and industry data. It displays coefficient tables, confidence-interval plots, and an expanding-window example.

What to notice

Check date alignment, missing windows and observation counts before interpreting a changing coefficient; a moving estimate is not evidence of an investable forecasting advantage.

Keep the context

Historical regression demonstration, with pointwise uncertainty. Live data downloads and package versions can change results. No realized investment performance or AI benefit is established.

AI use: Not reported in the source.

The inspected source does not report AI-assistant use by its authors.

A useful public example is not an assessment of a reader, a publisher or a Better Loop member.

Authored practice suggestion

Try the idea.
Check your own work.

Use material you are allowed to work with. This suggestion is preparation; it does not record a completed task or an improvement.

A prompt to adapt privately

Using a local snapshot of the linked factor and industry data, reconstruct the rolling regression. Show how dates align, where estimates first become available, and how missing observations affect a window. Compare one rolling estimate with an independently fitted regression on the same rows.

A check to adapt

The selected window and independent regression agree within a stated tolerance; initial missing estimates remain missing, and the risk-free adjustment is explicit.

Open a related practice brief

Private practice needs no account. Your configured model provider may process the evidence you select.

See the full skill-to-sharing flow

Another way into the work.

Partial pooling for Minnesota radon measurements

The notebook joins Minnesota household and county data, fits pooled, unpooled and hierarchical radon models, and displays posterior diagnostics and comparisons. The partial-pooling plots show how estimates change with county sample size.

Published byChris Fonnesbeck and contributors / PyMC

Curated public exampleGuide

Analysis & finance

Joining air-quality measurements to station metadata

The tutorial concatenates NO2 and particulate-measurement tables, shows row-count checks, and joins station coordinates and parameter descriptions. Displayed tables trace how the measurement dataset gains metadata.

Published bypandas documentation contributors