SAS-Makros

The %bms SAS macro selects variables for binomial logistic regression using a change in estimate algorithm. Potential confounders are selected based on the relative amount of change observed in estimated odds ratio of the independent variable of interest as each potential confounder is removed from the logistic model. The macro can be instructed to keep non-confounders based on crude improvement in the global model fit caused by the inclusion of the non-confounders. An option to create and examine all possible effect modifiers is also built into the macro.

Authors: Janice Hegewald, Prof. Dr. Annette Pfahlberg, Prof. Dr. Wolfgang Uter

License: GPL-2

Source code

Paper: „A Backwards Manual Selection Procedure for Binary Logistic Regression in the SAS ® PROC LOGISTIC Procedure.“ NESUG 16 Proceedings.

The SAS macros (RAP_LOGISTIC, RAP_GENMOD, RAP_PHREG) perform the point and interval estimates for a new epidemiological parameter, the risk/rate advancement period (RAP), derived from different generalized linear models, developed using SAS version 8.2.

Authors: Prof. Dr. Annette Pfahlberg, Tina van der Horst, Prof. Dr. Olaf Gefeller

License: GPL

Source code

Reference manual (in German)

Relevant papers:

  • Dempsey M (1947): Decline in tuberculosis; the death rate fails to tell the entire story. Am Rev Tuberculosis 86, 157.
  • Robins JM, Greenland S (1991): Estimability and estimation of expected years of life lost due to hazardous exposure. Stat Med 10, 79-93.
  • Brenner H, Gefeller O, Greenland S (1993): Risk and rate advancement periods as measures of exposure impact on the occurrence of chronic disease. Epidemiology 4, 229-236.
  • McCullagh P, Nelder JA: Generalized linear models. 2nd ed. Chapman & Hall, New York 1989.
  • Herson J (1975): Fieller’s theorem vs. delta method for significance intervals for ratios. J Statist Comput Simul 3, 265-274.
  • Fieller EC (1944): A fundamental formula in the statistics of biological assays and some applications. Q J Pharm Pharmacol 17, 117-123.
  • Quenouille M (1949): Approximation tests of correlation in time series. J Royal Statist Soc B 11, 18-44.
  • Tukey JW (1958): Bias and confidence in not quite large samples. (abstract) Ann Math Statist 29, 614.
  • Miller RG (1964): A trustworthy jackknife. Ann Math Statist 39, 1594-1605.
  • Efron B, Tibshirani RJ: An introduction to the bootstrap. Chapman & Hall, New York 1993.
  • Pfahlberg A, Gefeller O, Brenner H: Computational realization of point and interval estimation of risk and rate advancement periods. Epidemiol 1995;6:99-100.
  • Pfahlberg A. Statistische Aspekte des epidemiologischen Risikokonzepts der „Risk and Rate Advancement Period“. Fachbereich Statistik. Dortmund, Universität Dortmund, 1999.
  • Pfahlberg A, Gefeller O. Assessing the impact of classical risk factors on myocardial infarction by rate advancement periods. Am J Epidemiol 2001;154:486-488.