Page 261 - Contributed Paper Session (CPS) - Volume 7

CPS2084 Siti Aisyah Mohd Padzil et al.
paper will be focusing on obtaining the relationship between household
characteristic and state with poverty to examine whether or not it effects
among Malaysian’s household poverty.
To highlight the causes of household poverty, this paper will provide a
study on household income data and uses two modelling methods which are
Model Averaging and Alternative Model Averaging. Model Averaging is a well-
known statistical modelling approach which aims to solve issues of MS
regarding underestimation of parameter estimates. MA decreases the
estimates of a weaker variables by averaging weights of all possible model. In
other word, MA helps in making inferences based on the entire set of
candidate models (Posada and Buckley, 2004). Since MA will give best model
that includes all variables being studied, AMA is applied to shows a different
modelling approach. AMA approach is build based on MA approach except
for elimination of insignificant variables in the final model.

2. Methodology
2.1 Household Income Data
To point out the causes of poverty in Malaysia, data of Household Income
retrieved from Department of Statistic Malaysia was studied. The poverty level
is measured using net income and Malaysia’s PLI as in (Saidatulakmal, 2014).
The survey includes 13232 observations and nine independent variables as in
Table 1.

248 | I S I W S C 2 0 1 9