Filter results by

Search Help
Currently selected filters that can be removed

Keyword(s)

Type

1 facets displayed. 0 facets selected.

Content

1 facets displayed. 0 facets selected.
Sort Help
entries

Results

All (13)

All (13) (0 to 10 of 13 results)

  • Articles and reports: 12-001-X202300100010
    Description: Precise and unbiased estimates of response propensities (RPs) play a decisive role in the monitoring, analysis, and adaptation of data collection. In a fixed survey climate, those parameters are stable and their estimates ultimately converge when sufficient historic data is collected. In survey practice, however, response rates gradually vary in time. Understanding time-dependent variation in predicting response rates is key when adapting survey design. This paper illuminates time-dependent variation in response rates through multi-level time-series models. Reliable predictions can be generated by learning from historic time series and updating with new data in a Bayesian framework. As an illustrative case study, we focus on Web response rates in the Dutch Health Survey from 2014 to 2019.
    Release date: 2023-06-30

  • Articles and reports: 12-001-X201800154927
    Description:

    Benchmarking monthly or quarterly series to annual data is a common practice in many National Statistical Institutes. The benchmarking problem arises when time series data for the same target variable are measured at different frequencies and there is a need to remove discrepancies between the sums of the sub-annual values and their annual benchmarks. Several benchmarking methods are available in the literature. The Growth Rates Preservation (GRP) benchmarking procedure is often considered the best method. It is often claimed that this procedure is grounded on an ideal movement preservation principle. However, we show that there are important drawbacks to GRP, relevant for practical applications, that are unknown in the literature. Alternative benchmarking models will be considered that do not suffer from some of GRP’s side effects.

    Release date: 2018-06-21

  • Articles and reports: 12-001-X201700254871
    Description:

    In this paper the question is addressed how alternative data sources, such as administrative and social media data, can be used in the production of official statistics. Since most surveys at national statistical institutes are conducted repeatedly over time, a multivariate structural time series modelling approach is proposed to model the series observed by a repeated surveys with related series obtained from such alternative data sources. Generally, this improves the precision of the direct survey estimates by using sample information observed in preceding periods and information from related auxiliary series. This model also makes it possible to utilize the higher frequency of the social media to produce more precise estimates for the sample survey in real time at the moment that statistics for the social media become available but the sample data are not yet available. The concept of cointegration is applied to address the question to which extent the alternative series represent the same phenomena as the series observed with the repeated survey. The methodology is applied to the Dutch Consumer Confidence Survey and a sentiment index derived from social media.

    Release date: 2017-12-21

  • Articles and reports: 12-001-X201700114819
    Description:

    Structural time series models are a powerful technique for variance reduction in the framework of small area estimation (SAE) based on repeatedly conducted surveys. Statistics Netherlands implemented a structural time series model to produce monthly figures about the labour force with the Dutch Labour Force Survey (DLFS). Such models, however, contain unknown hyperparameters that have to be estimated before the Kalman filter can be launched to estimate state variables of the model. This paper describes a simulation aimed at studying the properties of hyperparameter estimators in the model. Simulating distributions of the hyperparameter estimators under different model specifications complements standard model diagnostics for state space models. Uncertainty around the model hyperparameters is another major issue. To account for hyperparameter uncertainty in the mean squared errors (MSE) estimates of the DLFS, several estimation approaches known in the literature are considered in a simulation. Apart from the MSE bias comparison, this paper also provides insight into the variances and MSEs of the MSE estimators considered.

    Release date: 2017-06-22

  • Articles and reports: 11-522-X20050019467
    Description:

    This paper reviews techniques for dealing with missing data from complex surveys when conducting longitudinal analysis. In addition to incurring the same types of missingness as cross sectional data, longitudinal observations also suffer from drop out missingness. For the purpose of analyzing longitudinal data, random effects models are most often used to account for the longitudinal nature of the data. However, there are difficulties in incorporating the complex design with typical multi-level models that are used in this type of longitudinal analysis, especially in the presence of drop-out missingness.

    Release date: 2007-03-02

  • Articles and reports: 12-001-X19990014709
    Description:

    We develop an approach to estimating variances for X-11 seasonal adjustments that recognizes the effects of sampling error and errors from forecast extension. In our approach, seasonal adjustment error in the central values of a sufficiently long series results only from the effect of the X-11 filtering on the sampling errors. Towards either end of the series, we also recognize the contribution to seasonal adjustment error from forecast and backcast errors. We extend the approach to produce variances of errors in X-11 trend estimates, and to recognize error in estimation of regression coefficients used to model, e.g., calendar effects. In empirical results, the contribution of sampling error often dominated the seasonal adjustment variances. Trend estimate variances, however, showed large increases at the ends of series due to the effects of fore/backcast error. Nonstationarities in the sampling errors produced striking patterns in the seasonal adjustment and trend estimate variances.

    Release date: 1999-10-08

  • Articles and reports: 12-001-X199600114383
    Description:

    The estimation of the trend-cycle with the X-11-ARIMA method is often done using the 13-term Henderson filter applied to seasonally adjusted data modified by extreme values. This filter however, produces a large number of unwanted ripples in the final or “historical” trend-cycle curve which are interpreted as false turning points. The use of a longer Henderson filter such as the 23-term is not an alternative for this filter is sluggish to detect turning points and consequently is not useful for current economic and business analysis. This paper proposes a new method that enables the use of the 13-term Henderson filter with the advantages of: (i) reducing the number of unwanted ripples; (ii) reducing the size of the revisions to preliminary values and (iii) no increase in the time lag to detect turning points. The results are illustrated with nine leading indicator series of the Canadian Composite Leading Index.

    Release date: 1996-06-14

  • Articles and reports: 12-001-X199100214505
    Description:

    The X-11-ARIMA seasonal adjustment method and the Census X-11 variant use a standard ANOVA-F-test to assess the presence of stable seasonality. This F-test is applied to a series consisting of estimated seasonals plus irregulars (residuals) which may be (and often are) autocorrelated, thus violating the basic assumption of the F-test. This limitation has long been known by producers of seasonally adjusted data and the nominal value of the F statistic has been rarely used as a criterion for seasonal adjustment. Instead, producers of seasonally adjusted data have used rules of thumb, such as, F equal to or greater than 7. This paper introduces an exact test which takes into account autocorrelated residuals following an SMA process of the (0, q) (0, Q)_s type. Comparisons of this modified F-test and the standard ANOVA test of X-11-ARIMA are made for a large number of Canadian socio-economic series.

    Release date: 1991-12-16

  • Articles and reports: 12-001-X199000214532
    Description:

    Births by census division are studied via graphs and maps for the province of Saskatchewan for the years 1986-87. The goal of the work is to see how births are related to time and geography by obtaining contour maps that display the birth phenomenon in a smooth fashion. A principal difficulty arising is that the data are aggregate. A secondary goal is to examine the extent to which the Poisson-lognormal can replace for data that are counts, the normal regression model for continuous variates. To this end a hierarchy of models for count-valued random variates are fit to the birth data by maximum likelihood. These models include: the simple Poisson, the Poisson with year and weekday effects and the Poisson-lognormal with year and weekday effects. The use of the Poisson-lognormal is motivated by the idea that important covariates are unavailable to include in the fitting. As the discussion indicates, the work is preliminary.

    Release date: 1990-12-14

  • Articles and reports: 12-001-X199000214533
    Description:

    A commonly used model for the analysis of time series models is the seasonal ARIMA model. However, the survey errors of the input data are usually ignored in the analysis. We show, through the use of state-space models with partially improper initial conditions, how to estimate the unknown parameters of this model using maximum likelihood methods. As well, the survey estimates can be smoothed using an empirical Bayes framework and model validation can be performed. We apply these techniques to an unemployment series from the Labour Force Survey.

    Release date: 1990-12-14
Data (0)

Data (0) (0 results)

No content available at this time.

Analysis (13)

Analysis (13) (0 to 10 of 13 results)

  • Articles and reports: 12-001-X202300100010
    Description: Precise and unbiased estimates of response propensities (RPs) play a decisive role in the monitoring, analysis, and adaptation of data collection. In a fixed survey climate, those parameters are stable and their estimates ultimately converge when sufficient historic data is collected. In survey practice, however, response rates gradually vary in time. Understanding time-dependent variation in predicting response rates is key when adapting survey design. This paper illuminates time-dependent variation in response rates through multi-level time-series models. Reliable predictions can be generated by learning from historic time series and updating with new data in a Bayesian framework. As an illustrative case study, we focus on Web response rates in the Dutch Health Survey from 2014 to 2019.
    Release date: 2023-06-30

  • Articles and reports: 12-001-X201800154927
    Description:

    Benchmarking monthly or quarterly series to annual data is a common practice in many National Statistical Institutes. The benchmarking problem arises when time series data for the same target variable are measured at different frequencies and there is a need to remove discrepancies between the sums of the sub-annual values and their annual benchmarks. Several benchmarking methods are available in the literature. The Growth Rates Preservation (GRP) benchmarking procedure is often considered the best method. It is often claimed that this procedure is grounded on an ideal movement preservation principle. However, we show that there are important drawbacks to GRP, relevant for practical applications, that are unknown in the literature. Alternative benchmarking models will be considered that do not suffer from some of GRP’s side effects.

    Release date: 2018-06-21

  • Articles and reports: 12-001-X201700254871
    Description:

    In this paper the question is addressed how alternative data sources, such as administrative and social media data, can be used in the production of official statistics. Since most surveys at national statistical institutes are conducted repeatedly over time, a multivariate structural time series modelling approach is proposed to model the series observed by a repeated surveys with related series obtained from such alternative data sources. Generally, this improves the precision of the direct survey estimates by using sample information observed in preceding periods and information from related auxiliary series. This model also makes it possible to utilize the higher frequency of the social media to produce more precise estimates for the sample survey in real time at the moment that statistics for the social media become available but the sample data are not yet available. The concept of cointegration is applied to address the question to which extent the alternative series represent the same phenomena as the series observed with the repeated survey. The methodology is applied to the Dutch Consumer Confidence Survey and a sentiment index derived from social media.

    Release date: 2017-12-21

  • Articles and reports: 12-001-X201700114819
    Description:

    Structural time series models are a powerful technique for variance reduction in the framework of small area estimation (SAE) based on repeatedly conducted surveys. Statistics Netherlands implemented a structural time series model to produce monthly figures about the labour force with the Dutch Labour Force Survey (DLFS). Such models, however, contain unknown hyperparameters that have to be estimated before the Kalman filter can be launched to estimate state variables of the model. This paper describes a simulation aimed at studying the properties of hyperparameter estimators in the model. Simulating distributions of the hyperparameter estimators under different model specifications complements standard model diagnostics for state space models. Uncertainty around the model hyperparameters is another major issue. To account for hyperparameter uncertainty in the mean squared errors (MSE) estimates of the DLFS, several estimation approaches known in the literature are considered in a simulation. Apart from the MSE bias comparison, this paper also provides insight into the variances and MSEs of the MSE estimators considered.

    Release date: 2017-06-22

  • Articles and reports: 11-522-X20050019467
    Description:

    This paper reviews techniques for dealing with missing data from complex surveys when conducting longitudinal analysis. In addition to incurring the same types of missingness as cross sectional data, longitudinal observations also suffer from drop out missingness. For the purpose of analyzing longitudinal data, random effects models are most often used to account for the longitudinal nature of the data. However, there are difficulties in incorporating the complex design with typical multi-level models that are used in this type of longitudinal analysis, especially in the presence of drop-out missingness.

    Release date: 2007-03-02

  • Articles and reports: 12-001-X19990014709
    Description:

    We develop an approach to estimating variances for X-11 seasonal adjustments that recognizes the effects of sampling error and errors from forecast extension. In our approach, seasonal adjustment error in the central values of a sufficiently long series results only from the effect of the X-11 filtering on the sampling errors. Towards either end of the series, we also recognize the contribution to seasonal adjustment error from forecast and backcast errors. We extend the approach to produce variances of errors in X-11 trend estimates, and to recognize error in estimation of regression coefficients used to model, e.g., calendar effects. In empirical results, the contribution of sampling error often dominated the seasonal adjustment variances. Trend estimate variances, however, showed large increases at the ends of series due to the effects of fore/backcast error. Nonstationarities in the sampling errors produced striking patterns in the seasonal adjustment and trend estimate variances.

    Release date: 1999-10-08

  • Articles and reports: 12-001-X199600114383
    Description:

    The estimation of the trend-cycle with the X-11-ARIMA method is often done using the 13-term Henderson filter applied to seasonally adjusted data modified by extreme values. This filter however, produces a large number of unwanted ripples in the final or “historical” trend-cycle curve which are interpreted as false turning points. The use of a longer Henderson filter such as the 23-term is not an alternative for this filter is sluggish to detect turning points and consequently is not useful for current economic and business analysis. This paper proposes a new method that enables the use of the 13-term Henderson filter with the advantages of: (i) reducing the number of unwanted ripples; (ii) reducing the size of the revisions to preliminary values and (iii) no increase in the time lag to detect turning points. The results are illustrated with nine leading indicator series of the Canadian Composite Leading Index.

    Release date: 1996-06-14

  • Articles and reports: 12-001-X199100214505
    Description:

    The X-11-ARIMA seasonal adjustment method and the Census X-11 variant use a standard ANOVA-F-test to assess the presence of stable seasonality. This F-test is applied to a series consisting of estimated seasonals plus irregulars (residuals) which may be (and often are) autocorrelated, thus violating the basic assumption of the F-test. This limitation has long been known by producers of seasonally adjusted data and the nominal value of the F statistic has been rarely used as a criterion for seasonal adjustment. Instead, producers of seasonally adjusted data have used rules of thumb, such as, F equal to or greater than 7. This paper introduces an exact test which takes into account autocorrelated residuals following an SMA process of the (0, q) (0, Q)_s type. Comparisons of this modified F-test and the standard ANOVA test of X-11-ARIMA are made for a large number of Canadian socio-economic series.

    Release date: 1991-12-16

  • Articles and reports: 12-001-X199000214532
    Description:

    Births by census division are studied via graphs and maps for the province of Saskatchewan for the years 1986-87. The goal of the work is to see how births are related to time and geography by obtaining contour maps that display the birth phenomenon in a smooth fashion. A principal difficulty arising is that the data are aggregate. A secondary goal is to examine the extent to which the Poisson-lognormal can replace for data that are counts, the normal regression model for continuous variates. To this end a hierarchy of models for count-valued random variates are fit to the birth data by maximum likelihood. These models include: the simple Poisson, the Poisson with year and weekday effects and the Poisson-lognormal with year and weekday effects. The use of the Poisson-lognormal is motivated by the idea that important covariates are unavailable to include in the fitting. As the discussion indicates, the work is preliminary.

    Release date: 1990-12-14

  • Articles and reports: 12-001-X199000214533
    Description:

    A commonly used model for the analysis of time series models is the seasonal ARIMA model. However, the survey errors of the input data are usually ignored in the analysis. We show, through the use of state-space models with partially improper initial conditions, how to estimate the unknown parameters of this model using maximum likelihood methods. As well, the survey estimates can be smoothed using an empirical Bayes framework and model validation can be performed. We apply these techniques to an unemployment series from the Labour Force Survey.

    Release date: 1990-12-14
Reference (0)

Reference (0) (0 results)

No content available at this time.

Date modified: