{smcl}
{* 15Apr2005}{...}
{hline}
help for {cmd:svypxcon}
{hline}

{title:Predicted means or proportions for a continuous X variable for survey data}   

{p 8 23 2} 
{cmdab:svypxcon} {it:yvar} [{cmd:if} {it:exp}] [{cmd:in} {it:range}]{cmd:,}
   {cmdab:x:var(}{it:xvar}{cmd:)}
   {cmdab:f:rom(}#{cmd:)}
   {cmdab:t:o(}#{cmd:)}
   {cmdab:i:nc(}#{cmd:)}
   [ {cmdab:adj:ust(}{it:covlist}{cmd:)} 
   {cmdab:p:oly(}#{cmd:)}
   {cmdab:cl:ass(}{it:classvar}{cmd:)}
   {cmdab:mod:el}
   {cmdab:sub:pop(}{it:subpop_spec}{cmd:)} 
   {cmdab:gr:aph}
   {cmdab:l:evel(}#{cmd:)}
   {cmdab:lin:ear} 
   {cmdab:nol:ist}
   {cmdab:sav:epred(}{it:filename}{cmd:)}
   {it:graph_options} ]


{title:Description}

{p 4 8 4}
{cmd:svypxcon} calculates and prints predicted values and 95% confidence intervals
   for linear or logistic model estimates for a continuous X variable corrected for
   the survey sampling scheme (weights), adjusted for covariates.  Default prints
   predicted values and confidence intervals; model estimates and graph can be shown.
   Optionally a quadradic or a quadratic and cubic term can be added to the model. An
   interaction between X and a nominal variable can be estimated and graphed. Dummy
   variables are created for the nominal variable with the lowest category defaulting
   to the reference group.  (Updated for Version 9)

{p 4 8 4}
{cmd:svypxcon} requires that the survey design variables be identified using {help svyset}


{title:Variables and options required}

{p 4}{it:yvar} is the dependent variable

{p 8 8 2}If {it:yvar} is continuous, defaults to linear regression

{p 8 8 2}If {it:yvar} is binary (0,1), defaults to logistic regression


{p 4 8 2}
{cmd:xvar(}{it:xvar}{cmd:)} -- continuous independent variable (interval or ordinal)

{p 4}{cmd:from(}#{cmd:)} -- bottom value for {it:xvar}

{p 4}{cmd:to(}#{cmd:)} -- top value for {it:xvar}

{p 4 8 2}
{cmd:inc(}#{cmd:)} -- increment size between bottom and top values (defaults to 1)


{title:Options} 

{p 4 8 2}   
{cmd:adjust(}{it:covlist}{cmd:)} lists any covariates. If none are specified,
   unadjusted means or proportions are reported. Covariates are set to their mean,
   based on observations used in the analysis, or can be set to user specified values
   (e.g., age=50 gender=1). (Note: if an {cmd:adjust} variable is nominal with more than
   2 categories, it must be defined with dummy variables in the {cmd:adjust} list,
   whereas dummy variables are created automatically in the {cmd:xvar} option)

{p 4 8 2}   
{cmd:poly(}2 or 3{cmd:)} -- polynomial terms added: 2=quadratic 3=quadradic and cubic

{p 4 8 2}
{cmd:class(}{it:classvar}{cmd:)} -- nominal variable for an {it:xvar} by 
   {it:classvar} interaction

{p 4 8 2}   
{cmd:model} -- for display purposes only, this option prints the regression table

{p 4 8 2}   
{cmd:subpop(}{it:subpop_spec}{cmd:)} -- specifies that estimates be computed for
    the single subpopulation identified in {it:subpop_spec}

{p 4 8 2}   
{cmd:graph} -- displays graph of predicted values and 95% confidence intervals;
    if {cmd:class()} is requested, confidence intervals are not displayed

{p 4 8 2}   
{cmd:level(}#{cmd:)} -- specifies the confidence level, in percent, for
   calculation of confidence intervals (default=95%)

{p 4 8 2}   
{cmd:linear} -- requests linear regression when {it:yvar} is binary (0,1); if not 
    specified, logistic regression is assumed

{p 4 8 2}   
{cmd:nolist} -- does not print list of predicted values and 95% CIs

{p 4 8 2}   
{cmd:savepred(}{it:filename}{cmd:)} -- saves adjusted values and CI's to a Stata file

    
{title:Examples}

{p 4 8 2}
{cmd:. svypxcon chol, xvar(age) from(20) to(80) inc(5) adjust(sys hra)}

{p 8 8 2}
Using linear regressin, calculates the predicted cholesterol values for
5-year increments of age from 20 years to 80 years (i.e., age=20,25,30,...,80)
adjusted for systolic blood pressure and heart rate; displays predicted values,
but does not display model or graph

{p 4 8 2}
{cmd:. svypxcon chol, xvar(sys) f(60) t(280) i(20) adj(hra) graph class(race)}

{p 8 8 2}
Using linear regression, calculates the predicted cholesterol values for
systolic blood pressure readings from 60 to 280 in 20mm/hg increments by
categories of race, adjusted for heart rate; tests for an age by race
interaction; displays graph and list of predicted values

{p 4 8 2}
{cmd:. svypxcon htn, xvar(chol) f(100) t(300) i(20) class(race) graph}

{p 8 8 2}
Using logistic regression, calculates the unadjusted probability of hypertension
for cholesterol from 100 to 300 in 20-unit increments, by categories of race; 
displays test for interaction between cholesterol and race; graphs the results

{p 4 8 2}
{cmd:. svypxcon htn, xvar(age) f(40) t(80) poly(3) adj(chol=250) graph}

{p 8 8 2}
Using logistic regression, calculates the probabilitity of hypertension for 
1-yr increments of age; terms for age-squared and age-cubed included in model
using the poly(3) option; also adjusted to a cholesterol level of 250; displays
graph


{title:Author}

{p 4 8 2}
{hi:J.M.Garrett}, Professor, School of Medicine, University of North Carolina,
  Chapel Hill, NC. 
  Email: {browse "mailto:joanne_garrett@med.unc.edu":joanne_garrett@med.unc.edu}