[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: xtgraph and ci2
I have just finished updating & revising xtgraph.
My thanks to Kit Baum for posting the new version at
Version 1.01 is now at
This update fixes a number of bugs:
* Options -median bar(no)- (recently noted on Statalist) now work
* File saving when using temporary file names is more reliable
* Missing values are correctly handled with stringtype groups
I have also
* Documented the important -model- option in the help file
* Added a -boxcox- option for compatibility with -model-
* Ensured compatibility with Stata 7.0, so allowing for a greater range of
Option -model- instructs xtgraph to use the most recent regression model
to estimate the average, (using predict), the standard deviation (predict,
or the standard error (predict, stdp). Confidence intervals and reference
are worked out in the usual way. This option works with most of Stata
many regression commands for continuous data. It can be used, for instance,
to produce graphs with pooled or robust standard errors, adjustments for
censored data, or autoregression. If the outcome has been transformed
fitting the model, the appropriate transfomation option (log, power, or
must be used.
This is the only ocasion where the distinction between options -power-
is important. All values are taken from the last fitted model, except for
frequency count used in working out confidence intervals. Models involving
log, logit or probit transformations (poisson, nbreg, logistic, xtprobit
etc.) are not
supported. If there is a demand, I may get round to including these.
In passing, a reminder that -ci2- was also updated to be compatibile with
recently. Now, I just need to do the work on -diagt-.
Paul T Seed (Paul.Seed@kcl.ac.uk)
Lecturer in Medical Statistics
Maternal & Fetal Research Unit,
Department of Obstetrics & Gynaecology,
Guy's Kings and St. Thomas' School of Medicine,
King's College London.
* For searches and help try: