5. Some simulations
Phillip S. Kott and Dan Liao
Previous | Next
Paralleling Kott
and Liao (2012), we generated a synthetic population,
of hospitals from the 2008 DAWN
public-use file. After creating
we independently drew 3,600
stratified simple random samples of size 400 from
using the strata definitions on
the public-use file. These definitions incorporate information on location and
hospital ownership (public or private) not directly provided on the file.
We set the stratum
sample sizes roughly proportional to a size measure
but never less than four. For
we used annual drug-related
emergency-room visits, which was always positive. The DAWN actually has a size
variable attached to every hospital in the frame: total emergency-room visits
in a previous year according to the American Hospital Association.
Unfortunately, it was not included on the public-use file. Design weights in
our simulations varied between 4.375 and 48, which allowed us to treat the
finite population correction factors as ignorable in variance estimation.
As in our original
paper, we generated a respondent sample
for each simulated sample based
on Bernoulli draw from the logistic function:
We also created alternative respondent samples
using
Both response
models produce unweighted overall response rates of around 54%, which is
similar to actual DAWN experience, where response is also a mildly increasing
function of the size variable. Notice that
is bounded even if neither
probability can be expressed by equation (2.4) with a finite
As in the previous
study, we focused on estimating population totals for three survey variables. Annual
drug-related emergency-room visits with adverse pharmaceutical reaction and
those resulting in deaths came from the public-use file. Since both these
variables were roughly linear in our size measure, the third “survey” variable
was artificially constructed. It was the size measure (annual drug-related
emergency-room visits) raised to the 1.3 power.
We investigated
eight estimators and estimates of their variance. These are summarized in Table
5.1. The first two featured calibration to the original sample only (equation (2.5)
with
with response assumed to be
logistic in the log of the size measure. That is to say, equation (2.3) was
employed with
The first estimator used
as the calibration vector while
the second used
which was more consistent with a
reasonable prediction model, at least for adverse reactions and deaths.
Our third and
fourth estimator featured calibration to the sample and population in a single
step (equation (2.5) with
and then
using
They were designed to be nearly
unbiased if either the logistic response model in
or the linear prediction model in
held.
Not surprisingly,
the (empirical) relative mean squared error of the fourth estimator is always
lower than the third. The reason is fairly obvious looking at equation (3.1)
and considering the consequence of
being 0 (calibration to the
population) rather than 1 (calibration to the sample).
The fifth through
eighth estimators were calibrated in two steps. The fifth and seventh
estimators employed the calibration weighting from the first estimator in its
first step, while the sixth and eighth employed the calibration weighting from
the second estimator. The fifth and sixth used
in their second step, while the
seventh and eighth were nearly pseudo-optimal (Kott 2011) using
and
in their second step. All four
employed the individual weight-adjustment functions:
As Kott (2011) showed these
are asymptotically identical to the
weight-adjustment function,
when
but prevent any
from falling below unity. Each is a version of
equation (4.1) with
and
Because the
nonresponse rate was so large, we did not encounter a problem computing the
third and fourth estimator using any of the simulated respondent samples. The
relative mean squared error of the fourth estimator was always slightly higher
than that of the seventh and eighth estimators, which incorporated nearly
pseudo-optimal calibration in their second step. Interestingly, this was not
the case when comparing the fourth estimator to the fifth and sixth estimators
which, although employing two steps, did not incorporate nearly pseudo-optimal
calibration.
Observe that
although the second estimator always had a smaller relative mean squared error
than the first, being more consistent with a reasonable prediction model (even
for
the survey variable appeared
closer to being linear in
than in
the other analogous pairs (fifth vs
sixth and seventh vs eighth) exhibited no clear pattern of superiority. This is
because it is the second-step residuals that are effectively modeled in
equation (4.4) not the
values.
Generating the
nonresponse with equation (5.2) than (5.1) did not seem to have much of an
impact on the results except for the relative biases of the first estimator. For
both adverse reactions and
the relative bias of this
estimator is over 40% of the relative mean squared error. That is likely
because both models that could be used to justify this estimator (response is
logistic in the log of the size measure and the survey variable is linear in
the log of the size measure) fail. Not surprisingly, since the relative bias is
such a large part of the relative mean squared error in these two situations,
underestimates mean squared error
badly. Nowhere else is the relative bias of
greater than 15%.
It seems that even
our artificial variable,
was close enough to being linear
in the size measure that bias was never an issue for any estimator other than
the first. The first estimator itself had a negligible relative bias when
response was a logistic model of the log of the size measure, as assumed.
Previous | Next