Agriculture Reference
In-Depth Information
Table 7.3 Relative efficiency of the sample mean (MSE/MSE
SRS
) for each design, estimated
using 10,000 replicated samples of the sparse population, for different sample sizes, trends, and
homogeneity.
No trend
Linear trend
Quadratic trend
Homogeneity
Homogeneity
Homogeneity
Design
n
Low
Med
High
Low
Med
High
Low
Med
High
GRTS
10
1.01
1.02
0.92
0.66
0.67
0.70
0.77
0.75
0.73
CUBE 1
10
1.01
1.01
0.98
0.57
0.56
0.63
0.99
0.99
0.99
CUBE 2
10
1.00
1.01
1.00
0.57
0.57
0.63
0.57
0.54
0.49
DUST 1
10
1.28
1.28
1.09
0.48
0.45
0.35
0.68
0.62
0.58
DUST 2
10
1.24
1.22
0.98
0.46
0.42
0.33
0.58
0.57
0.55
SCPS
10
1.01
1.01
0.88
0.63
0.62
0.62
0.77
0.75
0.72
LPM 1
10
1.00
1.00
0.90
0.61
0.62
0.63
0.77
0.75
0.72
LPM 2
10
1.01
1.01
0.89
0.63
0.62
0.63
0.76
0.74
0.72
GRTS
50
1.00
0.99
0.78
0.55
0.55
0.52
0.63
0.61
0.55
CUBE 1
50
1.01
1.02
0.98
0.51
0.52
0.59
1.01
1.02
1.00
CUBE 2
50
1.01
1.00
0.98
0.52
0.52
0.58
0.52
0.49
0.42
DUST 1
50
2.93
2.76
1.90
0.91
0.83
0.57
1.12
1.04
0.78
DUST 2
50
2.54
2.28
1.48
0.78
0.70
0.44
0.87
0.82
0.57
SCPS
50
1.01
1.00
0.69
0.52
0.52
0.44
0.58
0.55
0.45
LPM 1
50
1.02
0.99
0.70
0.52
0.52
0.45
0.60
0.56
0.45
LPM 2
50
1.02
1.00
0.70
0.53
0.52
0.44
0.59
0.55
0.45
GRTS
100
1.01
0.96
0.67
0.53
0.51
0.42
0.60
0.56
0.45
CUBE 1
100
0.99
1.00
0.99
0.51
0.51
0.60
1.01
1.00
1.00
CUBE 2
100
1.00
1.00
0.98
0.51
0.51
0.58
0.51
0.47
0.41
DUST 1
100
4.00
3.79
2.32
1.27
1.09
0.71
1.45
1.35
0.84
DUST 2
100
3.33
3.01
1.64
1.03
0.87
0.52
1.05
0.98
0.57
SCPS
100
1.01
0.97
0.63
0.51
0.51
0.39
0.55
0.51
0.38
LPM 1
100
1.02
0.97
0.63
0.52
0.50
0.38
0.55
0.51
0.38
LPM 2
100
1.03
0.96
0.63
0.52
0.50
0.39
0.56
0.51
0.38
execution time matters can be secondary to the choice of the design with respect to
its efficiency, but only if its outcome is obtained in a reasonable amount of time.
Looking at results obtained with
R
, it is clear that LPM (particularly the more
accurate version 1) and SCPS are the most computationally intensive of the
examined procedures. However, as with the CUBE algorithm, the number of
computational operations gradually increases with
n
and proportionally with
N
.
Thus, even if they are much slower than the other solutions, we can be confident
that they can effectively be applied to large spatial populations. They are only
limited by the amount of memory needed to store the distance matrix.
Note that the CPU time dramatically decreases when using
C
, and so these
comparisons become less remarkable.
Search WWH ::
Custom Search