Page 54 - Contributed Paper Session (CPS) - Volume 5

CPS861 Madeline Dumaua C.
the two procedures were compared based on their respective Relative Root
Mean Square Error (RRMSE), Coefficient of Variation (CV), and the ratio of the
absolute bias to the standard deviation. RRMSE was used to measure the
accuracy and precision of the design-based and synthetic estimates and CV
was used to measure reliability or “goodness” of the estimates. On the other
hand, the ratio of the absolute bias to the standard deviation was used to
indicate negligibility of the bias of estimates at 10%. The estimator with the
highest number of small values of RRMSE and CV was considered as the better
estimator to estimate the total number of farm equipment and facilities in this
study.
In the province of Bukidnon, estimating the total number of plow, harrow,
and sprayer is best represented by synthetic estimation. Estimates gained from
these characteristics are more accurate and precise than design-based
estimation. This can be attributed to the very strong correlation between
variables considered in this study are the auxiliary variable, which is the total
number of households. Moreover, these estimates are also reliable through
their biases are not negligible. This happens because theoretically synthetic
estimator is a biased estimator.
On the other hand, it is evident from the results that the total number of
plows, thresher and hand tractor are the best estimated by the design-based
estimator in the province of Marinduque, These parameters have weak and
moderate correlations with the auxiliary variable. Therefore, it is expected that
estimates produced by synthetic estimator are less accurate and precise.
In the province of Isabela, it can be seen in that synthetic estimator is more
preferred to be use in estimating the total number of plows, harrows, and
sprayers. The estimates are more accurate and precise than direct estimates.
Also, these estimates are more reliable. These parameters satisfied the
assumption that characteristic of the larger areas is same as that of small areas.
In Laguna, synthetic estimator is more preferred in estimating the total
number of sprayers. This characteristic satisfied the assumption that the
province ratio have same characteristic as that of the municipal ratio. All other
parameters of ownership are best estimated using design-based estimation.
Generally, the results of the estimation procedures employed in this study
suggest that design-based estimation has negligible bias compared to
synthetic estimation. In addition, design-based estimation produced more
accurate and precise estimates. Synthetic estimation failed to give greater
number of estimates which are accurate and precise because mainly of the
failure in the assumption of small area having same characteristics as that of a
larger area. Design-based estimation procedure considered in this study is
good enough to be used in estimating the total number of farm equipment
and facilities owned.

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