Testing for Integer Integration in Functional Time Series

A 22.4 MB collection of supplementary materials for a statistical testing procedure for functional time series. The methodology was applied to Canadian yield curve data and French sub-national age-specific mortality data, with findings suggesting integration orders of one or fractional. Author Won-Ki Seo published the materials on figshare under a CC-BY-4.0 license in June 2026.

Use Cases

• Testing for integer versus fractional integration orders in functional time series based on the described statistical procedure.
• Analyzing the stationarity properties of yield curve data based on the application to Canadian interest rates.
• Modeling long-term trends in age-specific mortality rates based on the application to French sub-national data.
• Benchmarking new statistical tests against the Monte Carlo simulation results included in the supplementary materials.

Strengths

• Includes supplementary materials for reproducing the work, as stated in the description.
• Applies a novel testing methodology to two distinct real-world domains: finance and demography.
• Released under a permissive CC-BY-4.0 license, allowing for reuse and modification.

Limitations

• The 22.4 MB size suggests a small dataset, potentially limiting the scope of included data.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for large-scale modeling.

Provenance

Source

figshare, author Won-Ki Seo.

Collection Method

Likely contains data and code from Monte Carlo simulations and applications to Canadian yield curves and French mortality data.

Freshness

Last updated 2026-06-04 16:09:23; freshness should be verified.

Geography

Canada and France, based on the described applications.