BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Is the difference in means always a good measure for an effect?
DTSTART;VALUE=DATE-TIME:20220627T130000Z
DTEND;VALUE=DATE-TIME:20220627T132000Z
DTSTAMP;VALUE=DATE-TIME:20220930T165415Z
UID:indico-contribution-293@conferences.enbis.org
DESCRIPTION:Speakers: Øystein Evandt (Evandt Industrial Statistics)\nWhen
comparing a medical treatment with placebo it is usual to apply a two-sam
ple t-test. n1 patients are given treatment and n2 patients are given plac
ebo. The standard assumptions for using a two-sample t-test are assumed. I
t is also assumed that large response values in the treatment group are de
sirable. Usually H0: “The distribution means are equal” is tested agai
nst H1: “The distribution means are different”. The distribution mean
difference\, which is to be estimated\, is used as a measure of how good t
he medical treatment is compared to placebo. This measure has a deficiency
. It is compatible with the situation that n1 and n2 are so large that the
test is statistically significant while large distribution standard devia
tions cause so large overlap of the data in the two groups that it is mean
ingless to denote the result as clinically significant. We propose the sup
plemental measure P(X1>X2)\, where X1 and X2 is respectively the response
of a randomly selected patient given the treatment and the response of a r
andomly selected patient given placebo. This measure can be interpreted as
an approximate fraction of patients in the population in question who wou
ld respond better to treatment than placebo. A formula for P(X1>X2)\, whic
h is a function of the unknown distribution means and the unknown distribu
tion standard deviations\, is derived. Based on this formula confidence in
tervals for P(X1>X2)\, with specified confidence level\, are constructed b
ased on parametric as well as non-parametric bootstrap\, where the specifi
ed confidence level is ascertained by double bootstrap.\n\nhttps://confere
nces.enbis.org/event/18/contributions/293/
LOCATION: EL2
URL:https://conferences.enbis.org/event/18/contributions/293/
END:VEVENT
END:VCALENDAR