Simulation-Based Multiple Testing for Multivariate Models with Financial Applications

Lynda Khalaf's research proposes bootstrap-based tests for non-nested hypotheses in multivariate regressions. The 1.0 MB dataset, last updated on 2026-04-21, includes simulation results and an application to asset pricing models using Fama and French factors as a null hypothesis against consumption-based and liquidity-augmented alternatives.

Use Cases

• Evaluating the size and power of bootstrap-based hypothesis tests based on the described simulation methodology
• Comparing asset pricing models using the Fama and French factors as a baseline against alternative factor sets
• Investigating the role of liquidity factors in model rejections over short sub-samples as described in the application

Strengths

• Includes simulation results documenting test properties for single and multiple alternatives
• Applies methodology to a concrete financial use case with named factor models (Fama and French, consumption-based, liquidity-augmented)
• Dataset is licensed under CC-BY-4.0, permitting broad reuse

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment
• Data may reflect the temporal and methodological bias inherent to the specific research simulation

Provenance

Source

Lynda Khalaf via figshare

Collection Method

Simulation-based research and application to financial data, likely using bootstrap methods

Time Range

Analysis conducted at a quarterly frequency over unspecified sub-samples

Freshness

Last updated 2026-04-21 16:58:52; freshness should be verified

Geography

null