Alternatively you can use -fracpred- and draw the graph yourself:
----------------------------------
sysuse auto, clear
fracpoly: regress price mpg weight
fracpred pred, for(mpg)
fracpred se, for(mpg) stdp
gen ub=pred + invnormal(0.975)*se
gen lb=pred - invnormal(0.975)*se
twoway rarea ub lb mpg, sort pstyle(ci) || line pred mpg, sort
----------------------------------
Phil
On 09/03/2013, at 8:31 AM, Scott Merryman <scott.merryman@gmail.com> wrote:
> You could edit out the -scatter- command (line numbers 113 to 122),
> change the program define statement and save the file as
> fracplot2.ado
>
> Scott
>
> On Fri, Mar 8, 2013 at 11:09 AM, Alexander Liddle
> <alexander.liddle@ndorms.ox.ac.uk> wrote:
>> Dear All,
>>
>> I'm afraid it's another Stata graphics question - this may all be very simple but I would appreciate any advice you might have.
>>
>> I am doing a survival study and the effect of my predictor on the survival hazard is non-linear. As a result, I'm looking at fractional polynomials using fracpoly and plotting them using fracplot:
>>
>> fracpoly: stcox var1 var2 var3
>> fracplot
>>
>> This gives me a nice fractional polynomial with scatters above and below it. I am particularly interested in the curve itself, and so suppress the markers by doing:
>>
>> fracplot, msymbol(i)
>>
>> Obviously this just removes the markers but the scatters are still there 'in spirit' and can be made visible using graph editor. What I would like to do is suppress them entirely and so allow the axes to shorten to just include the fractional polynomial curve - giving a similar appearance to the twoway graph fpfitci.
>>
>> Does anyone have any ideas?
>>
>> Thank you,
>>
>> Alex Liddle
>>
>>
>> *
>> * For searches and help try:
>> * http://www.stata.com/help.cgi?search
>> * http://www.stata.com/support/faqs/resources/statalist-faq/
>> * http://www.ats.ucla.edu/stat/stata/
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/faqs/resources/statalist-faq/
> * http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
