Page 251 - Special Topic Session (STS) - Volume 4
P. 251

STS583 Michael W. et al.

                      Target Value & Design  MSE Est. Pop. Mean CV% Deff n_psu n_ssu Est. Pop. Total
                Age PPS                     1.23    21.14   1.43 1.35  80  12    949363
                Age PPS (wrong size)        2.21    21.16   1.92 2.37  71  12    943549
                Age Random                  1.56    21.11    1.6 1.68  80  12    948448
                Age STR                     1.56    21.13   1.52 1.58  80  12    952152
                Age STRPPS                  1.34    21.11   1.37 1.27  80  12    952578
                Consumption PPS             0.91  563329.11  0.91 1.16  71  12   220150
                Consumption PPS (wrong size)  1.45  563365.35  1.19 2.29  71  12  219830
                Consumption Random          1.06  563828.89  1.05 1.59  71  12   219857
                Consumption STR             0.62  563819.86  1.01 1.48  71  12   219568
                Consumption STRPPS          0.52   564101.1  0.9 1.17  71  12    219749
                Employment Ratio PPS        3.57     0.45   4.26 3.21  71  12    950165
                Employment Ratio PPS (wrong size)  4.88  0.45  5.16 4.93  71  12  936439
                Employment Ratio Random     4.37     0.45   4.38 3.43  71  12    950630
                Employment STR              4.02     0.46   4.09 3.12  71  12    953019
                Employment STRPPS           3.36     0.45   3.91 2.76  71  12    951303
                Population Count PPS        1.38     NA      1.5  Inf  120  12   950962
                Population Count PPS (wrong size)  11.99  NA  8.93 Inf  120  12  946087
                Population Count Random     5.78     NA     6.14 Inf  120  12    951510
                Population Count STR        5.01     NA     5.54 Inf  120  12    951357
                Population Count STRPPS     1.44     NA     1.47 Inf  120  12    950941
                                       Table 1: Census based Approach

                Incorrect size measures have also been used as MOS to demonstrate the
            impact on the different estimates. The bigger the relationship with the target
            value, the larger is the impact of size on the MSE (in percent).
                Stratification has been applied by different variables as described in the
            appendix.
            1.1. Sampling  from  a  purely  satellite  based  sampling  frame,  where  the
                population values are derived from a raster of gridded population data.
                There are situations, where there is no frame at all. In this case the option
            exist  to  use  gridded  population  data.  In  our  simulation  we  have  used
            WorldPop, and results are shown bellow.


                  Target Value & Design  MSE Est. Pop. Mean CV% Deff n_psu Est. Pop. Total Av. Samp. Size Av. Wpop. Pop
              Age PPS                3.96    21.11  3.22 12.54 100  938156   6552      9833
              Age PPS (calibrated)   3.29    21.08  0.52 0.33 100  922565    6593      9962
              Consumption PPS        2.21  563961.95  1.8 8.97 100  215671   6598      9927
              Employment PPS         11.39   0.46   8.96 29.47 100  933408   6626      9948
              Employment PPS (calibrated)  8.68  0.46  0  0  100  965882     6630      10004
                                        Table 2: Worldpop only sample








                                                               240 | I S I   W S C   2 0 1 9
   246   247   248   249   250   251   252   253   254   255   256