I'm performing test on one variance and in order to asses normality of X, I've transformed data with first natural logarithm transformation,then square transformation and calculated confidence interval. I want to see as well how would confidence interval for not transformed data and I was just wondering how one should recalculate the confidence interval. I can of course calculate it with non transformed data, but I would like to know how to just retransform the confidence interval, that we already have.

  • 1
    $\begingroup$ What parameter are you estimating? A variance? A mean? Something else? How are computing its CI? $\endgroup$ – whuber Dec 16 '18 at 21:12
  • 2
    $\begingroup$ I'm estimating variance with chi square distribution. My confidence interval is [((n-1)*s^2)/upper critical value ; ((n-1)*s^2)/lower critical value] . I'm just computing it on the new transformed data. $\endgroup$ – Michał Glinicki Dec 17 '18 at 9:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.