Information Technology Reference
In-Depth Information
Fig. 4.
A Comparision of Return Rate
3.4
Different Time Span of Training Data as an Indicator
Bayesian probability is learning from experiment data and expert knowledge. Whether
the larger the learning data is the better? We choose the data from 2011.1 to 2011.4 as
check data, and the training data is N months data that one year before the check data.
That is, we first choose 2011.1 as the check data, then the three months training data is
from 2009.11 to 2010.1, the four months training data is from 2009.10 to 2010.1, and so
on. After doing all the experiments, we put the all three months' check data together,
and compare with the really data, the result is shown by Table2. From the table, the
different time span of training data doesn't make any difference. However, the table
shows that different Bayesian probabilities may make significant different.
Table 2.
Return Rate of Different Time Span of Training Data
Training
data
Return
Rate
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
3months
0.78
0.79
0.81
0.82
0.84
0.84
0.87
0.91
0.92
1.00
1.00
4months
0.78
0.82
0.82
0.82
0.84
0.84
0.86
0.87
0.89
1.00
1.00
5months
0.78
0.80
0.82
0.83
0.83
0.83
0.85
0.89
0.91
1.00
1.00
6months
0.78
0.80
0.81
0.82
0.82
0.84
0.86
0.88
0.90
1.00
1.00
7months
0.78
0.79
0.81
0.82
0.82
0.83
0.87
0.88
0.91
1.00
1.00
8months
0.78
0.79
0.81
0.83
0.84
0.87
0.87
0.90
0.91
1.00
1.00
9months
0.78
0.79
0.80
0.81
0.83
0.84
0.86
0.91
0.92
1.00
1.00
