To appear in Proceedings
of the conference on Seasonality, Seasonal Adjustment and their
implications for Short-Term Analysis and Forecasting
Timely detection of turning points: Should I use the seasonally
adjusted
or trend estimates?
Zuleika Menezes, Craig
H.
McLaren, Nick Von Sanden, Xichuan (Mark) Zhang, Melanie Black
Timely and accurate detection of turning points is an important issue in
analysing time series data. Different time series estimates, such as the
original estimates and the derived seasonally adjusted and trend-cycle
estimates, are available to help assess turning points. This paper focuses
on detection of turning points from time series estimates derived using a
univariate approach. We investigate the difference between using
seasonally adjusted and trend estimates for timely detection of time
points.
back
To appear in Proceedings
of the conference on Seasonality, Seasonal Adjustment and their
implications for Short-Term Analysis and Forecasting
SEASABS: Australian Bureau of Statistics seasonal adjustment package
Craig H. McLaren,
Duncan McCaskill, Xichuan (Mark) Zhang
SEASABS (SEASonal analysis, ABS standards) is a unique seasonal adjustment
package which uses a knowledge based system to aid and assist both expert
and non-expert users. This paper describes the seasonal adjustment
infrastructure, including SEASABS, and approaches currently used at the
Australian Bureau of Statistics. Future directions for seasonal adjustment
infrastructure within the Australian Bureau of Statistics are also
considered.
back
Statistica
Neerlandica
Rotation patterns and trend estimation for repeated surveys using rotation
group estimates
Craig H. McLaren and
David
G. Steel
A general approach for constructing filters to produce trend estimates
from a repeated survey is described. This approach accounts for the
correlation structure induced by the rotation pattern used in the survey.
Different filters are developed depending on whether the trend analysis is
based on elementary estimates available for each rotation group or overall
estimates obtained by combining the rotation group estimates. The
properties of trend estimates obtained directly from the elementary
estimates, those obtained from the simple average of the rotation group
estimates and trend estimates obtained from the best linear unbiased
estimates of the population characteristics of interest are compared.
These comparisons are done for a number of rotation patterns, enabling an
assessment of the impact of the choice of rotation patterns on trend
estimation.
back
Australian and
New Zealand Journal of Statistics
An Easter Proximity Effect: Modelling and Adjustment
Xichuan (Mark) Zhang,
Craig H. McLaren,
Caleb C. S.
Leung
The timing of Easter Sunday varies from one year to the next and can
affect time series data. To reveal the underlying movement of a time
series, the date of Easter's occurrence and its impact on the time series
have to be taken into account. New approaches are developed to model and
remove the impact of Easter. The monthly Australian Total Retail Turnover
series is used to illustrate the effectiveness of the modelling
approaches.
back
Journal of Official Statistics
The Effect of Different Rotation Patterns on the Revisions of Trend
Estimates
David
G. Steel and Craig H. McLaren
The X11 and X11ARIMA procedures are widely used to produce seasonally
adjusted and trend estimates from time series obtained from sample
surveys. The surveys are often based on designs in which there is sample
overlap between different periods. The degree of overlap is determined by
the pattern of inclusion of selected units over time, i.e., the rotation
pattern. An important issue in analysing the series is that trend
estimates at the end of the series are revised as estimatesfor recent
periods are added. This article considers the effects of different
rotation patterns on the mean squared error of the revisions of trend
estimates.
back
Survey
Methodolodgy
The Impact of Different Rotation Patterns on the Sampling Variance of
Seasonally Adjusted and Trend Estimates
Craig H. McLaren
and
David
G. Steel
Many economic and social time series are based on sample surveys which
have complex sample designs. The sample design affects the properties of
the time series. In particular, the overlap of the sample from period to
period affects the variability of the time series of survey estimates, and
the seasonally adjusted and trend estimates produced from them. The Census
X11 and X11ARIMA packages are commonly used to produce seasonally adjusted
estimates and can also be used to produce estimates of trend. This paper
considers the implications of different overlap patterns on the sampling
variance of seasonally adjusted and trend estimates obtained from time
series based on sample surveys.
back
University of
Wollongong, School of Mathematics and Applied Statistics
Designing Rotation Patterns and Filters for Trend Estimation in Repeated
Surveys
Craig H. McLaren
Many important economic and social time series are based on repeated
sample surveys. A key element in the design of a repeated sample survey is
the rotation pattern, which affects the sample overlap, and hence the
correlation between estimates at different lags. Seasonally adjusted and
trend estimates can be calculated to aid in the interpretation of the time
series. Most national statistical agencies use the X11 and X11ARIMA
seasonal adjustment packages. The rotation pattern affects the variability
of the time series of survey estimates and the trend and seasonally
adjusted estimates produced from them. This thesis considers the choice of
rotation pattern for trend estimation from a repeated survey.
The implications of different rotation patterns on the sampling variance
of seasonally adjusted and trend estimates is considered. An important
issue in analysing trend estimates is that estimates at the very end of a
time series are revised. The impact of different rotation patterns on the
mean square error of the revisions of trend estimates is assessed.
As well as altering the rotation pattern, the filters used to estimate
trend can be changed. Theory is developed to allow optimal trend filters
to be generated for series having a known correlation structure. This is
used to investigate rotation patterns and optimal filter combinations. The
use of a design means that the sample consists of a number of rotation
groups. In some cases it will be possible to obtain separate estimates for
each rotation group. Optimal trend estimates can then be found depending
on whether the individual rotation group or overall estimates are
available. The properties of trend estimates obtained directly from the
separate rotation group estimates, those obtained from the simple average
of the rotation group estimates and using best linear unbiased estimates
are compared for different rotation patterns.
back