Italian Mortality Data for Dynamic Generalized Additive Models, 2015-2020

Wei Zhang's dataset supports a study on variational inference for Bayesian dynamic generalized additive models applied to mortality analysis. The data includes Italian death counts from January 2015 to December 2020, with predictors and temporally dependent multivariate random effects. The associated files include PDF, RDATA, TXT, STAN, R, and CPP formats, totaling 2.1 MB.

Use Cases

• Forecasting mortality trends based on observed predictors and temporal random effects described in the model.
• Analyzing spatial dependence structures among geographical locations for cause-specific death counts.
• Evaluating the efficiency of variational inference algorithms compared to Markov chain Monte Carlo methods for high-dimensional time series.
• Studying the co-evolution of multiple causes of death over time using the described multivariate random effects component.

Strengths

• Dataset is licensed under CC-BY-4.0, permitting broad reuse with attribution.
• The study applies a novel variational inference scheme claimed to be more efficient than several competing methods.
• Data covers a 6-year time range (January 2015 to December 2020) for Italy.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Wei Zhang via figshare.

Collection Method

Likely collected from official Italian mortality statistics, as described in the study.

Time Range

January 2015 to December 2020.

Freshness

Last updated 2026-05-14 18:56:37; freshness should be verified.

Geography

Italy.