scholarly journals Evaluation of Generalized Variance Functions in the Analysis of Complex Survey Data

2014 ◽  
Vol 30 (1) ◽  
pp. 63-90 ◽  
Author(s):  
MoonJung Cho ◽  
John L. Eltinge ◽  
Julie Gershunskaya ◽  
Larry Huff
Keyword(s):  
Mean Squared Error ◽  
Sample Survey ◽  
Complex Survey ◽  
Generalized Variance ◽  
Variance Estimators ◽  
Variance Functions ◽  
Complex Survey Data ◽  
Standard Design ◽  
Sample Data ◽  
Generalized Variance Functions

Abstract Two sets of diagnostics are presented to evaluate the properties of generalized variance functions (GVFs) for a given sample survey. The first set uses test statistics for the coefficients of multiple regression forms of GVF models. The second set uses smoothed estimators of the mean squared error (MSE) of GVF-based variance estimators. The smooth version of the MSE estimator can provide a useful measure of the performance of a GVF estimator, relative to the variance of a standard design-based variance estimator. Some of the proposed methods are applied to sample data from the Current Employment Statistics survey.

scholarly journals Better Coverage Intervals for Estimators from a Complex Sample Survey

2020 ◽  
Author(s):  
Phillip S. Kott
Keyword(s):  
Normal Distribution ◽  
Parameter Estimate ◽  
Regression Coefficients ◽  
Sample Survey ◽  
Complex Survey ◽  
Asymptotically Normal ◽  
Complex Survey Data ◽  
Complex Sample ◽  
Chi Squared ◽  
Coverage Interval

Coverage intervals for a parameter estimate computed using complex survey data are often constructed by assuming the parameter estimate has an asymptotically normal distribution and the measure of the estimator’s variance is roughly chi-squared. The size of the sample and the nature of the parameter being estimated render this conventional “Wald” methodology dubious in many applications. I developed a revised method of coverage-interval construction that “speeds up the asymptotics” by incorporating an estimated measure of skewness. I discuss how skewness-adjusted intervals can be computed for ratios, differences between domain means, and regression coefficients.

On normal stable Tweedie models and power-generalized variance functions of only one component

Test ◽  
2014 ◽  
Vol 23 (3) ◽  
pp. 585-606 ◽  
Author(s):  
Yacouba Boubacar Maïnassara ◽  
Célestin C. Kokonendji
Keyword(s):  
Generalized Variance ◽  
Variance Functions ◽  
Generalized Variance Functions

scholarly journals Estimation of multi-state life table functions and their variability from complex survey data using the SPACE Program

2010 ◽  
Vol 22 ◽  
pp. 129-158 ◽  
Author(s):  
Liming Cai ◽  
Mark Hayward ◽  
Yasuhiko Saito ◽  
James Lubitz ◽  
Aaron Hagedorn ◽  
...  
Keyword(s):  
Life Table ◽  
Survey Data ◽  
Space Program ◽  
Complex Survey ◽  
Complex Survey Data

scholarly journals Small Area Estimation of Postfire Tree Density Using Continuous Forest Inventory Data

2021 ◽  
Vol 4 ◽  
Author(s):  
George C. Gaines ◽  
David L. R. Affleck
Keyword(s):  
Forest Inventory ◽  
Mean Squared Error ◽  
Tree Density ◽  
Standard Errors ◽  
Sample Sizes ◽  
Management Planning ◽  
Forest Management Planning ◽  
Continuous Forest ◽  
Sample Data ◽  
Regional Forest

Wildfire activity in the western United States is expanding and many western forests are struggling to regenerate postfire. Accurate estimates of forest regeneration following wildfire are critical for postfire forest management planning and monitoring forest dynamics. National or regional forest inventory programs can provide vegetation data for direct spatiotemporal domain estimation of postfire tree density, but samples within domains of administrative utility may be small (or empty). Indirect domain expansion estimators, which borrow extra-domain sample data to increase precision of domain estimates, offer a possible alternative. This research evaluates domain sample sizes and direct estimates in domains spanning large geographic extents and ranging from 1 to 10 years in temporal scope. In aggregate, domain sample sizes prove too small and standard errors of direct estimates too high. We subsequently compare two indirect estimators—one generated by averaging over observations that are proximate in space, the other by averaging over observations that are proximate in time—on the basis of estimated standard error. We also present a new estimator of the mean squared error (MSE) of indirect domain estimators which accounts for covariance between direct and indirect domain estimates. Borrowing sample data from within the geographic extents of our domains, but from an expanded set of measurement years, proves to be the superior strategy for augmenting domain sample sizes to reduce domain standard errors in this application. However, MSE estimates prove too frequently negative and highly variable for operational utility in this context, even when averaged over multiple proximate domains.

Sign in / Sign up

Export Citation Format

Share Document