7. Conclusions and discussion
Paul Knottnerus
Previous
This section summarizes a number of conclusions and issues for further research.
When totals of
turnover are estimated from a panel in months
and
two estimators
and
for the growth rate between these
months can be distinguished.
When using
one should be aware that in
practice,
might be much smaller than
especially when the turnover in
month
and the turnover in month
are highly correlated and the
overlap ratios
and
are not too small.
The efficiency of
and
can be improved by the composite
estimator
described in Section 4.
Using least
squares techniques, an aligned composite vector-estimator
can be derived that obeys the
nonlinear restriction for totals and growth rates:
The AC estimates estimator
subject to linear restrictions can be
extended in several ways: (i) for nonlinear restrictions, (ii) for different
data sets such as monthly, quarterly and yearly data, (iii) for births and
deaths, (iv) for regression and ratio estimators, and (v) for additional
auxiliary variables.
Similar to the
regression estimator, the AC estimates estimator is asymptotically unbiased. This remark
also applies to the covariance-matrix estimator
There is not yet
an unambiguous answer on the question of to what extent data from the past
should be included in the vector estimate
each month. The answer depends
upon: (i) the NSI’s policy and rules with respect to revision of already
published figures, (ii) the fact that from a theoretical viewpoint, the
sequence of
monthly SRS estimates
(included as component in
should be so long that the
difference between the two AC estimates estimators of
say
and
is not substantial, and (iii) the
size of the samples. That is, in analogy with the regression estimator or,
equivalently, the calibration estimator, the sample sizes should be much larger
than the number of (calibration) restrictions. For a simulation study on the
variance of the regression estimator and the number of regressors, see Silva
and Skinner (1997) and for a relationship between the regression estimator and
the GR estimator, see Appendix A.3 and Knottnerus (2003).
In the specific
case of estimating mutually aligned totals and changes, additional research is
needed for finding: (i) the optimal and practical length for the monthly,
quarterly, semesterly and yearly series of SRS estimates to be included in the
initial vector
and (ii) a rule of thumb with
respect to the number of restrictions compared to the sample sizes in order to
find an AC estimates estimator
with an improved efficiency.
Acknowledgements
The
views expressed in this paper are those of the author and do not necessarily
reflect the policy of Statistics Netherlands. The author would like to thank
Harm Jan Boonstra, Arnout van Delden, Sander Scholtus, the Associate Editor and
two anonymous referees for their helpful comments and corrections.
Appendix
A.1 Proofs of (3.4) and (4.5)
The proof of (3.4)
is as follows. For
formula (2.2) can be rewritten as
Dividing (2.4) by (A.1) yields
In the last line we used that under the model assumptions mentioned in Section
3,
and
provided that
is sufficiently large; see also
the derivation of (3.2).
Next, under the
same assumptions, (4.5) can be derived as follows. Since
the covariance in (4.3) can be
rewritten as
Combining (2.4), (A.1) and (A.3), we can write
in (4.2) as
Similar to deriving (A.2), we used in the last line
A.2 Derivation of (5.5)
In case of
linear restrictions
matrix
can be found by minimizing
see Knottnerus (2003, page 330). The solution of this least squares
problem is given by
In case of
nonlinear restrictions, the new
minimand is
Similarly to (A.4), it can be shown that this minimand attains its
minimum for
Substituting Taylor’s linearization
into (A.5), we get the following
approximation, say
for
Assuming that
the first approximation for the
constrained maximum likelihood (ML) solution, say
can be calculated in the standard
manner by using the linearized restrictions
where
is defined by (5.4). If
does not satisfy the nonlinear
restrictions
a better approximation of
might be obtained by replacing
in (A.6) by update
resulting in a new matrix
In turn, in analogy with (A.7)
leads to a better approximation
or update of
say
where we used Taylor’s linearization of the nonlinear restrictions
around
Repeating this procedure, we get
the following recursions for
or, for short,
For definitions of
and
see Section 5; in practice,
should be replaced by its
estimate
By construction, for each
we have
see (5.4). Hence, when
converges to the (constrained)
maximum likelihood solution
converges to zero. Also, assuming
converges to say
the corresponding covariance
matrix of
say
can be approximated by
which for sufficiently large
can be estimated by
see also Cramer (1986, page 38).
A.3 Regression estimator as GR estimator
Suppose that
and the auxiliary variable
with known population mean
are observed in
In order to apply the GR
estimator to this situation, define
The prior restriction is
Applying (5.1) and (5.2) to this case yields the following GR estimator
Hence, replacing
by its estimate
we can approximate the first
element in
by
which corresponds to the familiar
regression estimator, often denoted by
For sufficiently large
the variance of
can be approximated by
recall from regression theory that
and
The variance in (A.8) can be
estimated by the well-known variance estimator
Similar results can be derived for more than one auxiliary variable.
This illustrates once more that with respect to the bias and the variance
approximation the AC estimates estimator strongly resembles the regression estimator or,
equivalently, the calibration estimator.
References
Berger, Y.G. (2004). Variance estimation for measures of
change in probability sampling. The
Canadian Journal of Statistics, 32, 451-467.
Cochran, W.G. (1977). Sampling Techniques. New York: John Wiley and Sons, Inc.
Cramer, J.S. (1986). Econometric
Applications of Maximum Likelihood Methods. Cambridge: Cambridge University
Press.
Hidiroglou, M.A., Särndal, C.E. and Binder, D.A. (1995).
Weighting and estimation in business surveys. In Business Survey Methods, (Eds., B.G. Cox et al.). New York: John
Wiley and Sons, Inc.
Kalman, R.E. (1960). A new approach to linear filtering
and prediction problems. Transactions ASME, Journal
of Basic Engineering, 82, 35-45.
Kish, L. (1965). Survey
sampling. New York: John Wiley and Sons, Inc.
Knottnerus, P. (2003). Sample Survey Theory: Some Pythagorean Perspectives. New York:
Springer-Verlag.
Knottnerus, P. and Van Delden, A. (2012). On variances
of changes estimated from rotating panels and dynamic strata. Survey Methodology, 38(1), 43-52.
Laniel, N. (1987). Variances for a rotating sample from
a changing population. Proceedings of the Survey Research Methods Section, American
Statistical Association, 496-500.
Magnus, J.R. and Neudecker, H. (1988). Matrix Differential Calculus with
Applications in Statistics and Econometrics. New York: John Wiley and Sons,
Inc.
Nordberg, L. (2000). On variance estimation for measures
of change when samples are coordinated by the use of permanent random numbers. Journal of Official Statistics, 16,
363-378.
Qualité, L. and Tillé, Y. (2008). Variance estimation of
changes in repeated surveys and its application to the Swiss survey of value
added. Survey Methodology, 34(2),
173-181.
Särndal, C.E., Swensson, B. and Wretman, J. (1992). Model Assisted Survey Sampling. New
York: Springer-Verlag.
Silva, P.L.D.N. and Skinner, C.J. (1997). Variable
selection for regression estimation in finite populations. Survey Methodology, 23(1), 23-32.
Smith, P., Pont, M. and Jones, T. (2003). Developments in business survey methodology in the Office for National Statistics, 1994–2000. Journal of the Royal Statistical Society D, 52, 257–295.
Tam, S.M. (1984). On covariances from overlapping samples. The American Statistician, 38, 288–289.
Wood, J. (2008). On the covariance between related
Horvitz-Thompson estimators. Journal of Official Statistics, 24, 53-78.
Previous