Page 215 - Contributed Paper Session (CPS) - Volume 6

CPS1892 Yang Xinhong
Table 2 "NOF" unit number in relevant strata and the sample size

In the practical survey, sample replacement happens when sample unit
refuses to obey, the data provided is undoubtedly false, or contacts cannot
be reached. The principle of sample replacement is to extract the next one of
the isometric sampling position of the original sample, excluding the
following situations:(1) samples without "NOF" activities during the reporting
period (the "NOF" amount is 0); (2) samples of extinction (closing down) (the
"NOF" amount is treated as 0); (3) samples having moved out of Shenzhen
(the "NOF" amount is treated as 0).
2.2.2 Parameter Estimation
Point estimation-The total "NOF" amount of self-employed households
in the whole city is obtained by point estimation, and the data are obtained
by adding the parameters of the "big self-employed households" stratum and
the "general self-employed households" stratum. The detailed calculation
formula is shown in Table 3.

Table 3 "NOF" population amount and calculation formula for parameter in each stratum

Interval estimation-The interval estimation (95% confidence level Z
distribution) of the total "NOF" amount of the self-employed households in
Shenzhen can be calculated by the following formula:
̂
̂
+ 0.025 . √( ) (1)

̂
̂
− 0.025 . √( ) (2)

The population variance of "NOF" amount of self-employed households
and "big self-employed households" in Shenzhen can be calculated by the
following fonnula.

( ) = ∑ ( ) (3)
̂
̂
ℎ

ℎ=1

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