EUVIMEDEUVIMED entdecken
Crossref

Specification testing with complex survey data

Mengmeng Xu, Puying Zhao, Niansheng Tang

Electronic Journal of Statistics · 01.01.2026 · 20 (2) · 5353-5392

DOI: 10.1214/26-ejs2607 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronic Journal of Statistics, erschienen am 01.01.2026. Verfasst von Mengmeng Xu und 2 weiteren Autor:innen. Die Arbeit nennt 55 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Model specification testing has received extensive attention. Existing approaches, however, largely assume independent and identically distributed observations within model-based frameworks and are often ill-suited to the dependence and weighting structures of complex survey data. We develop a novel nonparametric specification test for linear quantile models with complex survey data using a design-based approach. Our framework uses an orthogonal projection onto the tangent space of nuisance parameters to assess the correct specification of a linear conditional quantile function over a continuum of quantile levels. The proposed test statistic serves two key purposes: it enables the assessment of model validity for design-based inference regarding the conditional effects on outcome distributions; and it effectively accommodates the survey design features and provides valid inference for both the particular finite population and the related superpopulation. We derive the asymptotic distribution of the test statistic under the null and alternative hypotheses and show that orthogonality removes the first-order effect of parameter estimation. This orthogonality facilitates critical value approximation using a simple multiplier bootstrap procedure and simplifies the proof of its theoretical validity. Simulations and an empirical application demonstrate the advantages of the proposed method.

Zitieren

APA

Xu, M., Zhao, P., & Tang, N. (2026). Specification testing with complex survey data. Electronic Journal of Statistics, 20(2), 5353-5392. https://doi.org/10.1214/26-ejs2607

Vancouver

Xu M, Zhao P, Tang N. Specification testing with complex survey data. Electronic Journal of Statistics. 2026;20(2):5353-5392. doi:10.1214/26-ejs2607

BibTeX
@article{Xu2026,
  author = {Xu, Mengmeng and Zhao, Puying and Tang, Niansheng},
  title = {Specification testing with complex survey data},
  journal = {Electronic Journal of Statistics},
  year = {2026},
  volume = {20},
  number = {2},
  pages = {5353--5392},
  doi = {10.1214/26-ejs2607}
}
Kostenlos starten →