Hi all, I wanted to graph the predicted probabilities and confidence
intervals of a dependent variable named "scale" for different values of
two independent variables: "hmo" and "ppo". The model includes hmo
squared, ppo squared and an interaction term of the two. The following
commands work when it comes to graphing the predicted probabilities over
the 0.1 to 0.6 range of the hmo and ppo variables. However, I want to
know how to plot the CI's. praccum reports everything but I don't know
how to save the CI's and plot them. Any suggestions? Any suggestions on
how to plot marginal effects? Thanks, Jose Pagan.
tobit scale [[set of X's]] hmo hmo2 ppo ppo2 hmoppo, ul(1)
* HMO GRAPH
capture matrix drop mage
forvalues count = 0.1(0.1)0.6 {
local countsq = `count'*`count'
local countppo = `count'*ppo
prvalue, x(hmo `count' hmo2 `countsq' hmoppo `countppo') rest(mean)
brief
praccum, using(mage) xis(`count')
}
praccum, using(mage) gen(hmsq)
scatter hmsqxb hmsqx || qfit hmsqxb hmsqx
* PPO GRAPH
capture matrix drop ppoe
forvalues pcount = 0.1(0.1)0.6 {
local pcountsq = `pcount'*`pcount'
local pcounthmo = `pcount'*hmo
prvalue, x(ppo `pcount' ppo2 `pcountsq' hmoppo `pcounthmo')
rest(mean) brief
praccum, using(ppoe) xis(`pcount')
}
praccum, using(ppoe) gen(pposq)
scatter pposqxb pposqx || qfit pposqxb pposqx
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
