I'm using the user-written - stpm2 - written by Royston and Lambert. The commands are also available in the book Flexible Parametric Survival Analysis Using Stata.
Precisely at the page 152, the authors recommend to install the ado file named str2d in order to get the D statistics (for discrimination) as well as the model's R-squared.
I installed the ado file without problems, and checked its help files. So far, so good.
However, if I try to reproduce the same command line shown at page 152 (with the freely available rott2 dataset), I get this error message:
Code:
. xi: stpm2 enodes_1 i.size grade pr_1 chemo hormon age, scale(odds) df(2) i.size _Isize_1-3 (naturally coded; _Isize_1 omitted) Iteration 0: log likelihood = -3344.467 Iteration 1: log likelihood = -3330.7033 Iteration 2: log likelihood = -3330.6376 Iteration 3: log likelihood = -3330.6375 Log likelihood = -3330.6375 Number of obs = 2,982 ------------------------------------------------------------------------------ | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- xb | enodes_1 | -2.469988 .1262644 -19.56 0.000 -2.717462 -2.222515 _Isize_2 | .3873098 .0786168 4.93 0.000 .2332237 .5413958 _Isize_3 | .6630704 .1288065 5.15 0.000 .4106143 .9155265 grade | .4909539 .0853156 5.75 0.000 .3237384 .6581693 pr_1 | -.0831007 .0164349 -5.06 0.000 -.1153125 -.0508888 chemo | -.7075086 .102732 -6.89 0.000 -.9088597 -.5061576 hormon | -.6668108 .1200258 -5.56 0.000 -.902057 -.4315647 age | -.0190175 .0031803 -5.98 0.000 -.0252507 -.0127843 _rcs1 | 1.287639 .0292011 44.10 0.000 1.230406 1.344872 _rcs2 | .2205794 .0208198 10.59 0.000 .1797733 .2613854 _cons | .885118 .3346087 2.65 0.008 .2292969 1.540939 ------------------------------------------------------------------------------ . xi: str2d stpm2 enodes_1 i.size grade pr_1 chemo hormon age, scale(odds) df(2) i.size _Isize_1-3 (naturally coded; _Isize_1 omitted) unsupported or unrecognised command, stpm2 r(198);
To no avail.
One thing is for sure: the command stpm is supported (since I could perform the survival analysis). Also, the str2d was correctly downloaded (since its help file worked perfectly).
I wonder whether you can help me to solve this issue.
Thanks in advance
0 Response to Flexible Parametric Survival (stpm2) - command str2d not working
Post a Comment