Display options
Cite
Bliss RY, Weinberg J, Vieira V, et al. Power of permutation tests using generalized additive models with bivariate smoothers. J Biom Biostat. 2010;1(104)doi: 10.4172/2155-6180.1000104.
Bliss, R. Y., Weinberg, J., Vieira, V., Ozonoff, A., & Webster, T. F. (2010). Power of permutation tests using generalized additive models with bivariate smoothers. Journal of biometrics & biostatistics, 1(104), . https://doi.org/10.4172/2155-6180.1000104
Bliss, Robin Y, et al. "Power of permutation tests using generalized additive models with bivariate smoothers." Journal of biometrics & biostatistics vol. 1,104 (2010). doi: https://doi.org/10.4172/2155-6180.1000104
Bliss RY, Weinberg J, Vieira V, Ozonoff A, Webster TF. Power of permutation tests using generalized additive models with bivariate smoothers. J Biom Biostat. 2010 Sep 12;1(104). doi: 10.4172/2155-6180.1000104. PMID: 21701694; PMCID: PMC3119353.
Copy Download .nbib Format: NLM
Share it on

J Biom Biostat. 2010 Sep 12;1(104). doi: 10.4172/2155-6180.1000104.

Power of permutation tests using generalized additive models with bivariate smoothers.

Journal of biometrics & biostatistics

Robin Y Bliss, Janice Weinberg, Veronica Vieira, Al Ozonoff, Thomas F Webster

Affiliations

  1. Department of Biostatistics, Boston University School of Public Health.

PMID: 21701694 PMCID: PMC3119353 DOI: 10.4172/2155-6180.1000104

Abstract

In spatial epidemiology, when applying Generalized Additive Models (GAMs) with a bivariate locally weighted regression smooth over longitude and latitude, a natural hypothesis is whether location is associated with an outcome. An approximate chi-square test (ACST) is available but has an inflated type I error rate. Permutation tests provide alternatives. This research evaluated powers of ACST and four permutation tests: the conditional (CPT), fixed span (FSPT), fixed multiple span (FMSPT), and unconditional (UPT) permutation tests. For CPT, the span size was determined by minimizing the Akaike Information Criterion (AIC) and was held constant for models applied to permuted datasets. For FSPT, a single span was selected a priori. For FMSPT, GAMs were applied using 3-5 different spans selected a priori and the significance cutoff was reduced to account for multiple testing. For UPT, the span was selected by minimizing the AIC for observed and for permuted datasets. Data with a cluster of increased/decreased risk centered in a study region were simulated. ACST and CPT had high power estimates when applied with reduced significance cutoffs to adjust for inflated type I errors. FSPT power depended on the span size; FMSPT power estimates were slightly lower. Overall, UPT had low power.

References

  1. Environ Health. 2005 Jun 14;4:11 - PubMed
  2. Int J Health Geogr. 2010 Jul 19;9:37 - PubMed
  3. Stat Med. 2005 Mar 15;24(5):753-73 - PubMed
  4. Environ Res. 1997 Feb;72(2):118-30 - PubMed
  5. Int J Health Geogr. 2007 Nov 27;6:52 - PubMed
  6. Environ Health. 2005 Sep 15;4:19 - PubMed
  7. Am J Prev Med. 2006 Feb;30(2 Suppl):S37-49 - PubMed
  8. Int J Health Geogr. 2005 May 18;4:11 - PubMed
  9. Comput Stat Data Anal. 2011 Jan 1;55(1):366-374 - PubMed
  10. Stat Med. 1995 Apr 30;14(8):799-810 - PubMed
  11. Int J Health Geogr. 2010 Feb 12;9:7 - PubMed
  12. Int J Health Geogr. 2003 Dec 19;2(1):9 - PubMed
  13. Int J Health Geogr. 2006 Jun 09;5:26 - PubMed
  14. Environ Health. 2009 Feb 10;8:3 - PubMed

Publication Types

Grant support