.-
help for ^rocsize^ (version 1.0.0)
.-
Power for one sample screening study - continuous data
------------------------------------------------------
^rocsize^ tp_rate fp_rate, ^nd(^#^)^ ^ndb(^#^)^ ^tpn^ull^(^#^)^ ^fpn^ull^(^#^)^
[^,^ ^a^lpha^(^#^)^ ^ns^im^(^#^)^ ^d^ots ^nodis^play ^kern^al^(^kernal name^)^
^kwid^th^(^#^)^ ^res^file^(^filename^)^ ^replace^]
Description
-----------
^rocsize^ determines power to detect ROC curve improvement relative to the
binormal ROC defined by the joint fixed null true (^tpnull(^#^)^) and false
positive (^fpnull(^#^)^) rates for a sample drawn from a population with
binormal ROC defined specified alternative rates (^tp_rate^ & ^fp_rate^) and
specified sample sizes for diseased (^nd(^#^)^) and non-diseased (^ndb(^#^)^)
subjects.
Power is based on rejection of the null (tpr_0, fpr_0) relative to the
one-sided upper confidence bound for the fpr(at tpr_0) and/or lower bound
for the tpr(at fpr_0) for simulated samples.
Null and alternative true and false positive rates should be specified as
numbers between zero and one.
returned results (power):
r(pow_w1) based on Wald interval rejection; roc(fpr_0) lower bound > tpr_0
r(pow_w2) " " " " " fpr(tpr_0) upper bound < fpr_0
r(pow_w3) " on rejection per either of the above
r(pow_l1), r(pow_l2), r(pow_l3) as above, but using logit(roc) based
confidence interval rejection
Options
-------
^nsim(^#^)^ specifies the number of simulated study samples to be drawn for
determination of power. The default is 500 samples if not otherwise
specified.
^alpha(^#^)^ specifies the type I error rate. the default is
1 - ($S_level/100.), i.e. .05 if the level has not been set. See @level@
^nodisplay^ suppresses output display (in case where only returned results are
desired).
^dots^ optionally displays a dot on the screen for each simulated sample drawn.
^kernal(^kernal name^)^ specifies the kernal option for the kernal density
calculation used in computing the confidence bands. This is passed to the
kdensity command. see help ^kdensity^. By default the Epanechnikov density
is used.
^kwidth(^#^)^ specifies the halfwidth of the kernal, the width of the density window
around each point. See ^[R] kdensity^ regarding the default "optimal width".
^resfile(^filename^)^ creates a Stata file (^.dta^ file) with the simulation
results.
^replace^ requests that if an existing file is specified for saving simulation
results it should be overwritten.
Remarks
-------
Persons interested in reproducibility of results should set the random-number
seed by typing set seed # before running ^scrtest^; see help @generate@.
Simulated samples generated under the alternative (tpr_1, fpr_1) hypothesis are
drawn from
Y_dbar ~ N(0,1) and Y_d ~ N(mu,1)
where
mu = PHI_inv(1-fpr_1) + PHI_inv(tpr_1)
Author
------
Gary Longton, Fred Hutchinson Cancer Research Ctr.
glongton@@fhcrc.org
Also see
--------
On line: help for @roctab@, @emroc@ if installed,
@scrsize@ if installed, @aucsize@ if installed,