var_est_mvrg

Run variance estimation using the multivariate regression approach for proportional 0/1 indicator.

%var_est_mvrg(idsn=, odsn=, yr=, cty_var=, strt=, clstr=, wght=, prp_ind=, ilib=WORK, olib=WORK);

Arguments

• idsn : the name of the input dataset in the work directory; default: not set;
• yr : the year for which the variance calculated;
• cty_var : the name of the column in the input dataset which contains the country codes;
• strt : the name of the column in the input dataset which contains the starta information;
• clstr : the name of the column in the input dataset which contains the cluster/PSU information;
• wght : the name of the column in the input dataset which contains the weight information;
• prp_ind : the name of the column in the input dataset which contains the indicator variable which contains 0 or 1;
• ilib : (option) input library; default: ilib is set to WORK.

Returns

• odsn : a generic string used for the naming of output datasets; namely, the following datasets are created:
  • outdata_&odsn : output dataset containig all the data is needed to estimate the standard error for an absolute change between two cross-sectional estimators in diffrent years;
  • result_&odsn_total : output dataset containig the year, the country, the value 1 (as the value where the indicator takes the value 1), the value of the indicator and the standard error;
• olib : (option) output library; default: olib is set to WORK.

Example

Run for instance: ~~~sas var_est_mvrg(idsn = adat_prep, odsn = mvrg, yr = 2014, cty_var= DB020, strt = DB050, clstr = DB030, wght = RB050a, prp_ind = arope); ~~~

Note

The macro %var_mvrg_cmpr uses the implementation of variance estimation developed within the context of NET-SILC2 by [Osier et al, 2013] following the original algorithm of [Berger and Priam, 2010 and 2016].

References

1. Atkinson B., Guio A.-C. and Marlier E. eds (2017): "Monitoring social inclusion in Europe".
2. Berger, Y.G. and Priam, R. (2016): "A simple variance estimator of change for rotating repeated surveys: an application to the EU-SILC household surveys".
3. Osier G., Berger Y. and Goedeme T. (2013): "Standard error estimation for the EU–SILC indicators of poverty and social exclusion".
4. Berger Y. and Priam R. (2010): "Estimation of correlations between cross-sectional estimates from repeated surveys: an application to the variance of change".
5. Atkinson B. and Marlier E. eds (2010): "Income and living conditions in Europe".

See also

%var_est_data_prep, %var_mvrg_cmpr, %var_est_srvyfrq.