Stata 15 help for estat svy

[SVY] estat -- Postestimation statistics for survey data

Syntax

Survey design characteristics

estat svyset

Design and misspecification effects for point estimates

estat effects [, estat_effects_options]

Design and misspecification effects for linear combinations of point estimates

estat lceffects exp [, estat_lceffects_options]

Subpopulation sizes

estat size [, estat_size_options]

Subpopulation standard-deviation estimates

estat sd [, estat_sd_options]

Singleton and certainty strata

estat strata

Coefficient of variation for survey data

estat cv [, estat_cv_options]

Goodness-of-fit test for binary response models using survey data

estat gof [if] [in] [, estat_gof_options]

Display covariance matrix estimates

estat vce [, estat_vce_options]

estat_effects_options Description ------------------------------------------------------------------------- deff report DEFF design effects deft report DEFT design effects srssubpop report design effects, assuming SRS within subpopulation meff report MEFF design effects meft report MEFT design effects display_options control spacing and display of omitted variables and base and empty cells -------------------------------------------------------------------------

estat_lceffects_options Description ------------------------------------------------------------------------- deff report DEFF design effects deft report DEFT design effects srssubpop report design effects, assuming SRS within subpopulation meff report MEFF design effects meft report MEFT design effects -------------------------------------------------------------------------

estat_size_options Description ------------------------------------------------------------------------- obs report number of observations (within subpopulation) size report subpopulation sizes -------------------------------------------------------------------------

estat_sd_options Description ------------------------------------------------------------------------- variance report subpopulation variances instead of standard deviations srssubpop report standard deviations, assuming SRS within subpopulation -------------------------------------------------------------------------

estat_cv_options Description ------------------------------------------------------------------------- nolegend suppress the table legend display_options control spacing and display of omitted variables and base and empty cells -------------------------------------------------------------------------

estat_gof_options Description ------------------------------------------------------------------------- group(#) compute test statistic using # quantiles total compute test statistic using the total estimator instead of the mean estimator all execute test for all observations in the data -------------------------------------------------------------------------

Menu

Statistics > Survey data analysis > DEFF, MEFF, and other statistics

Description

estat svyset reports the survey design characteristics associated with the current estimation results.

estat effects displays a table of design and misspecification effects for each estimated parameter.

estat lceffects displays a table of design and misspecification effects for a user-specified linear combination of the parameter estimates.

estat size displays a table of sample and subpopulation sizes for each estimated subpopulation mean, proportion, ratio, or total. This command is available only after svy: mean, svy: proportion, svy: ratio, and svy: total; see [R] mean, [R] proportion, [R] ratio, and [R] total.

estat sd reports subpopulation standard deviations based on the estimation results from mean and svy: mean; see [R] mean. estat sd is not appropriate with estimation results that used direct standardization or poststratification.

estat strata displays a table of the number of singleton and certainty strata within each sampling stage. The variance scaling factors are also displayed for estimation results where singleunit(scaled) was svyset.

estat cv reports the coefficient of variation (CV) for each coefficient in the current estimation results. The CV for coefficient b is

CV(b) = SE(b)/|b| x 100%

estat gof reports a goodness-of-fit test for binary response models using survey data. This command is available only after svy: logistic, svy: logit, and svy: probit; see [R] logistic, [R] logit, and [R] probit.

estat vce displays the covariance or correlation matrix of the parameter estimates of the previous model. See [R] estat vce for examples.

Options for estat effects

deff and deft request that the design-effect measures DEFF and DEFT be displayed. This is the default, unless direct standardization or poststratification was used.

The deff and deft options are not allowed with estimation results that used direct standardization or poststratification. These methods obscure the measure of design effect because they adjust the frequency distribution of the target population.

srssubpop requests that DEFF and DEFT be computed using an estimate of simple random sampling (SRS) variance for sampling within a subpopulation. By default, DEFF and DEFT are computed using an estimate of the SRS variance for sampling from the entire population. Typically, srssubpop is used when computing subpopulation estimates by strata or by groups of strata.

meff and meft request that the misspecification-effect measures MEFF and MEFT be displayed.

display_options: noomitted, vsquish, noemptycells, baselevels, allbaselevels; see [R] estimation options.

Options for estat lceffects

deff and deft request that the design-effect measures DEFF and DEFT be displayed. This is the default, unless direct standardization or poststratification was used.

The deff and deft options are not allowed with estimation results that used direct standardization or poststratification. These methods obscure the measure of design effect because they adjust the frequency distribution of the target population.

srssubpop requests that DEFF and DEFT be computed using an estimate of simple random sampling (SRS) variance for sampling within a subpopulation. By default, DEFF and DEFT are computed using an estimate of the SRS variance for sampling from the entire population. Typically, srssubpop is used when computing subpopulation estimates by strata or by groups of strata.

meff and meft request that the misspecification-effect measures MEFF and MEFT be displayed.

Options for estat size

obs requests that the number of observations used to compute the estimate be displayed for each row of estimates.

size requests that the estimate of the subpopulation size be displayed for each row of estimates. The subpopulation size estimate equals the sum of the weights for those observations in the estimation sample that are also in the specified subpopulation. The estimated population size is reported when a subpopulation is not specified.

Options for estat sd

variance requests that the subpopulation variance be displayed instead of the standard deviation.

srssubpop requests that the standard deviation be computed using an estimate of SRS variance for sampling within a subpopulation. By default, the standard deviation is computed using an estimate of the SRS variance for sampling from the entire population. Typically, srssubpop is used when computing subpopulation estimates by strata or by groups of strata.

Options for estat cv

nolegend prevents the table legend identifying the subpopulations from being displayed.

display_options: noomitted, vsquish, noemptycells, baselevels, allbaselevels; see [R] estimation options.

Options for estat gof

group(#) specifies the number of quantiles to be used to group the data for the goodness-of-fit test. The minimum allowed value is group(2). The maximum allowed value is group(df), where df is the design degrees of freedom (e(df_r)). The default is group(10).

total requests that the goodness-of-fit test statistic be computed using the total estimator instead of the mean estimator.

all requests that the goodness-of-fit test statistic be computed for all observations in the data, ignoring any if or in restrictions specified with the model fit.

Options for estat vce

See estat_vce_options.

Examples

--------------------------------------------------------------------------- . webuse nhanes2 . svy: mean tcresult tgresult . estat svyset . estat effects, deff deft meff meft . estat size

. svy: mean tcresult, over(sex) . estat size . estat sd . estat cv

--------------------------------------------------------------------------- Design effects for subpopulations

. webuse nhanes2b . svy: mean iron, over(sex) . estat effects . estat effects, srssubpop

. webuse nmihs . svy: mean birthwgt, over(race) . estat effects . estat effects, srssubpop

--------------------------------------------------------------------------- Misspecification effects for subpopulations

. webuse nhanes2b . svy: mean zinc, over(sex) . estat effects, meff meft

--------------------------------------------------------------------------- Design and misspecification effects for linear combinations

. webuse nhanes2b . svy: mean tcresult, over(sex) . estat lceffects [tcresult]Male - [tcresult]Female, deff deft meff meft

--------------------------------------------------------------------------- Using survey data to determine Neyman allocation

. webuse nmihs . svyset [pw=finwgt], strata(stratan) . svy: mean birthwgt, over(stratan)

. estat size . matrix p_obs = 100 * r(_N)/e(N) . matrix nsubp = r(_N_subp)

. estat sd . matrix p_neyman = 100 * hadamard(nsubp,r(sd))/el(nsubp*r(sd)',1,1) . matrix list p_obs, format(%4.1f) . matrix list p_neyman, format(%4.1f) ---------------------------------------------------------------------------

Stored results

estat svyset stores the following in r():

Scalars r(stages) number of sampling stages r(stages_wt) last stage containing stage-level weights

Macros r(wtype) weight type r(wexp) weight expression r(wvar) weight variable name r(weight#) variable identifying weight for stage # r(su#) variable identifying sampling units for stage # r(strata#) variable identifying strata for stage # r(fpc#) FPC for stage # r(bsrweight) bsrweight() variable list r(bsn) bootstrap mean-weight adjustment r(brrweight) brrweight() variable list r(fay) Fay's adjustment r(jkrweight) jkrweight() variable list r(sdrweight) sdrweight() variable list r(sdrfpc) fpc() value from within sdrweight() r(vce) vcetype specified in vce() r(dof) dof() value r(mse) mse, if specified r(poststrata) poststrata() variable r(postweight) postweight() variable r(rake) rake() specification r(regress) regress() specification r(settings) svyset arguments to reproduce the current settings r(singleunit) singleunit() setting

estat strata stores the following in r():

Matrices r(_N_strata_single) number of strata with one sampling unit r(_N_strata_certain) number of certainty strata r(_N_strata) number of strata r(scale) variance scale factors used when singleunit(scaled) is svyset

estat effects stores the following in r():

Matrices r(deff) vector of DEFF estimates r(deft) vector of DEFT estimates r(deffsub) vector of DEFF estimates for srssubpop r(deftsub) vector of DEFT estimates for srssubpop r(meff) vector of MEFF estimates r(meft) vector of MEFT estimates

estat lceffects stores the following in r():

Scalars r(estimate) point estimate r(se) estimate of standard error r(df) degrees of freedom r(deff) DEFF estimate r(deft) DEFT estimate r(deffsub) DEFF estimate for srssubpop r(deftsub) DEFT estimate for srssubpop r(meff) MEFF estimate r(meft) MEFT estimate

estat size stores the following in r():

Matrices r(_N) vector of numbers of nonmissing observations r(_N_subp) vector of subpopulation size estimates

estat sd stores the following in r():

Macros r(srssubpop) srssubpop, if specified

Matrices r(mean) vector of subpopulation mean estimates r(sd) vector of subpopulation standard-deviation estimates r(variance) vector of subpopulation variance estimates

estat cv stores the following in r():

Matrices r(b) estimates r(se) standard errors of the estimates r(cv) coefficients of variation of the estimates

estat gof stores the following in r():

Scalars r(p) p-value associated with the test statistic r(F) F statistic, if e(df_r) was stored by estimation command r(df1) numerator degrees of freedom for F statistic r(df2) denominator degrees of freedom for F statistic r(chi2) chi-squared statistic, if e(df_r) was not stored by estimation command r(df) degrees of freedom for chi-squared statistic

estat vce stores the following in r():

Matrices r(V) VCE or correlation matrix


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index