Airline new customer tier level forecasting for real-time resource allocation of a miles program

Journal of Big Data

Table 7 Comparison of prediction power of extrapolation vs. D/S model

Question asked D months after 1st flight: will they become *Silver* within S months?
Case	D	S	Extrapolation of miles Rx¹	D/S Model Rx²P³A⁴
0	0.5	3	0.39	0.60 81% 31%
1	1	1	0.39	0.71 87% 53%
2	1	2	0.39	0.70 89% 48%
3	3	1	0.50	0.89 97% 82%⁵
4	3	3	0.39	0.67 95% 46%
5	6	3	0.51	0.83 96% 69%

Notes. ¹To compare predictive power between two models we use correlation as a proxy. Rx = correlation of accumulated miles with binary variable silver_attain. ²correlation of the Model’s prediction with binary variable silver_attain. ³P = Precision, of all users who were predicted to become Silver, How many of them did indeed become silver? ⁴A = Accuracy: Of all users who will become silver how many did we not miss? 5: See Table 8 for details on case 3.