30
Stata Technical Bulletin
STB-58
Example 4
Using the same model, we request the probability of coronary heart disease for a 40 year old who does not smoke, exercises
regularly, and has a cholesterol value of 200.
. predcalc chd, xvar(age=40 smk=O exer=l chl=200)
Model: Logistic Regression
Outcome: Coronary heart disease — chd
X Values: age=40 smk=0 exer=l chl=200
Mum. Obs: 609
Predicted Value and 95% CI for chd:
0.0249 (0.0118, 0.0518)
For the individual with no strong risk factors, the probability of developing coronary heart disease is only 0.02.
References
Cassel, J. C. 1971. Summary of major findings of the Evans County heart disease study. Archives of Internal Medicine 128(8): 887-889.
snp15.2 UpdatetoSomersd
Roger Newson, Guy’s, King’s and St Thomas’ School of Medicine, London, UK, [email protected]
Abstract: somersd calculates confidence intervals for rank order statistics. It has been updated to handle long variable lists.
Keywords: Somers’ D, Kendall’s tau, rank correlation, confidence intervals, nonparametric methods.
The command somersd introduced in Newson (2000a) and updated in Newson (2000b) has again been updated, this time
to handle long variable lists (it was previously limited to lists of 8 variables). It has also been improved, streamlined, debugged,
and intensively certified.
References
Newson, R. 2000a. snp15: somersd - Confidence limits for nonparametric statistics and their differences. Stata Technical Bulletin 55: 47-55.
——. 2000b. snp15.1: Update to somersd. Stata Technical Bulletin 57: 35.
snp16 Robust confidence intervals for median and other percentile differences between two groups
Roger Newson, Guy’s, King’s and St Thomas’ School of Medicine, London, UK, [email protected]
Abstract: A program is presented for calculating robust confidence intervals for median (and other percentile) differences (and
ratios) between values of a variable in two samples. The median difference is the same as that produced by the program
c id, using the Conover method. However, the confidence limits are typically different, being robust to the possibility that
the two population distributions differ in ways other than location, such as having unequal variances. The program uses
somersd.
Keywords: robust, confidence interval, median, percentile, difference, ratio, rank-sum, Wilcoxon, two-sample.
Syntax
cendif depvar [using filename] Weeihht] [if exp] [in range], hy(groupvar) [ centileUnumlist)
level(#) eform cluster (varuame) tdist transf({z ∣ asin ∣ iden})
saving (filename [. replace] ) nohold ]
fweights, iweights and pweights are allowed. They are treated as described in Methods and formulas below.
Description
cendif calculates confidence intervals for median differences, and other percentile differences, between values of a Y-
variable in depvar for a pair of observations chosen at random from two groups A and B, defined by the groiipvar in the by
option. These confidence intervals are robust to the possibility that the population distributions in the two groups are different
in ways other than location. This might happen if, for example, the two populations had different variances. For positive-valued
variables, cendif can be used to calculate confidence intervals for median ratios or other percentile ratios. cendif requires the
program somersd from Newson (2000).