Page 116 - Contributed Paper Session (CPS) - Volume 7
P. 116

CPS2043 Adnan Dawood K. B. et al.
               dwelling),  dwelling  characteristics  (such  as  the  number  of  bedrooms,
               availability  of  some  durable  goods),  demographic  characteristics  (such  as
               household  size,  gender  of  the  head  of  the  household),  education
               characteristics  for  the  household  members,  health  characteristics,  and  the
               labor  force  characteristics  for  the  household  members.  These  suggested
               independent  variables  are  used  in  constructing  the  PMTF  to  predict  the
               dependent variable; per capita income/ expenditure. The predicted income/
               expenditure (per capita) from the PMTF is used to compare with the household
               poverty line. The household is classified as poor when the predicted per capita
               income  from  the PMTF  model  is  less  than  the poverty  line; otherwise, the
               household  is  classified  as  non-poor.  The  weakness  of  the  PMTF  is  always
               overestimated for the poorest segment of the households (our desired target);
               especially when the national poverty incidence is low.
               The aim of this paper is to enhance the efficiency of the PMTF through building
               more than one formula to target the poor households. The main idea is to
               select three cutoff points on the natural logarithm of the annual per capita
               income/ expenditure. These cutoff points should be critical as much as we can,
               to  separate  the  population  into  four  segments  of  households:  the  first
               segment  is  the  High  Welfare  Level  (HWL)  which  don’t  include  any  poor
               household, the second is the Middle Welfare Level (MWL) which could include
               poor households, the third segment is the Lower Middle Welfare Level (LMWL)
               which  also  could  include  poor  households,  and  the  fourth  is  the  Bottom
               Welfare Level (BWL) which includes most of the poor households. Three PMTFs
               are  required  to  classify  the  population  on  these  four  categories.  The
               households which belongs to the HWL will be ignored because they are not
               including  any  poor  household.  For  the  remaining  three  categories  of
               households, we need to build another three specific PMTFs, formula for each
               category. The range of the first PMTF is lying between the first cutoff point
               and the second cutoff point and will represent the MWL. The range for the
               second PMTF is lying between the second cutoff point and the third cutoff
               point  and  will  represent  the  LMWL.  The  range  of  the  third  PMTF  is  lying
               between the third cutoff point and the origin point and will represent the BWL.
               Figure (1) and Figure (2) explain the idea:










                                                                  103 | I S I   W S C   2 0 1 9
   111   112   113   114   115   116   117   118   119   120   121