Stata Programs: Sampling Uncertainty for CostEffectiveness (Revised 05/03/13)

IPROGS.DO: "Immediate form" programs to estimate Fieller's method confidence intervals for
costeffectiveness ratios (fielleri), confidence intervals for NMB (nmb1i and nmbi), acceptability
curves (accept1i and accepti), and value of information curves. Also included is a program that defines the CI below which the lower limit is
less than the upper limit; the CI between which the lower limit is greater than the upper limit;
and the largest defined (ciboundi). Third, there is a program that calculates the inputs for these
four immediate form programs (ipinputs). Finally, iprogsdoc provides brief documentation of the
programs. All programs are included in iprog.do. Three sample datasets, exper1.dta, exper2.dta,
and exper3.dta are provided. For exper1, the lower 95% limit is less than the upper 95% limit;
for exper2, the lower 95% limit is greater than the upper 95% limit is greater than the point
estimate; and for exper3, the 95% CI is undefined. Finally, a .log file is provided which demonstrates
use of the programs. Programs particularly useful if one has estimated costs and outcomes by use
of tests other than the ttests.

ceagraphs.do: Programs that when used in conjunction with fielleri, accepti, and nmbi graphically display the point estimate and confidence limits for the costeffectiveness ratio (fiellergraph), the acceptability graph (accgraph), the point estimate and confidence limits for NMB defined for varying levels of WTP (nmbgraph), and the value of information curve. The ceagraphsdoc program provides a brief overview of the programs and the syntax for running them.
 ceagraphs.do: Stata .do file (revised 5/06/11; works with revised iprogs.do only)
 Data
Lecture notes related to measuring sampling uncertainty for economic analysis
 UPROGS.DO: Programs that use ttests to estimate Fieller's method confidence intervals for costeffectiveness
ratios (fielleru), confidence intervals for NMB (nmbu), and the acceptability curves (acceptu). Also included is a
program that defines the CI below which the lower limit is less than the upper limit; the CI between which the
lower limit is greater than the upper limit; and the largest defined (ciboundu). Finally, uprogsdoc provides brief
documentation of the programs. All programs are included in uprog.do. Three sample datasets, exper1.dta,
exper2.dta, and exper3.dta are provided. For exper1, the lower 95% limit is less than the upper 95% limit; for
exper2, the lower 95% limit is greater than the upper 95% limit is greater than the point estimate; and for exper3,
the 95% CI is undefined. Finally, a .log file is provided which demonstrates use of the programs. Programs useful
if one wants to estimate costs and outcomes by use of ttests.
Programs to estimate confidence intervals for costeffectiveness ratios (bscicer), confidence intervals for
NMB (bsnmb), and the acceptability curve (bsaccept) from bootstrapped estimates of cost and outcome. All 3
programs are included in bsceaprogs.do. They presume you have already constructed the bootstrap sample. Three
sample datasets, bsexper1.dta, bsexper2.dta, and bsexper3.dta, are provided. In addition, a program for
constructing a bootstrap sample (by use of simple differences in means) is provided (bscer.do). Finally , a .log
file is provided which demonstrates use of bscicer. bsnmb amd bsaccept.
 bsceagraphs.do: Programs that when used in conjunction with bscicer, bsnmb,bsaccept, and bsvoi graphically display the point estimate and confidence limits for the costeffectiveness ratio (bscicergraph), the acceptability graph (bsaccgraph), and the point estimate and confidence limits for NMB defined for varying levels of WTP (bsnmbgraph). The bsceagraphsdoc program provides a brief overview of the programs and the syntax for running them
 bsceagraphs.do: Stata .do file ( works with revised bsceaprogs.do only)
 Data
 Lecture notes related to bootstrapping confidence intervals for oosteffectiveness ratio
Updated 11/7/2005
