Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Risk Ratios and Risk Difference using SVY and jackknife

    Hi all,

    I'm trying to get risk ratios and risk differences for several analyses I'm doing with complex survey data, using the jackknife method for variance estimation. I found the on the Norton "Computing adjusted risk ratios and risk differences in Stata", describing their ADJRR command, which works with Taylor's series, but doesn't work with Jackknife.

    Has anyone figured out how to calculate risk ratios and risk difference using the svy command and jackknife for variance?

    Much appreciated!!

    Megan


  • #2
    to start, see
    Code:
    help svy jackknife
    risk differences are computed using regression; you can either use the -regress- command or -glm- with the identity link; similarly, for risk ratios, you can use -glm- with a log link, or, possibly, use -poisson- (but I'm not sure that the robust SE's you need for poisson can be used with -svy jackknife-)

    Comment

    Working...
    X