Thinking outside the curve, part I: modeling birthweight distribution

Table 1 Preferences of Model Selection Criteria on Real Data

Number of Components	FLIC 5000	BIC 5000	AIC 5000	FLIC 10000	BIC 10000	AIC 10000
1	0	0	0	0	0	0
2	20	21	1	8	11	0
3	5	4	4	5	5	0
4	0	0	14	12	9	15
5	0	0	1	0	0	1
6	0	0	4	0	0	8
7	0	0	1	0	0	1
Number of Components	FLIC 25000	BIC 25000	AIC 25000	FLIC 50000	BIC 50000	AIC 50000
1	0	0	0	0	0	0
2	0	0	0	0	0	0
3	0	0	0	0	0	0
4	25	25	5	22	22	1
5	0	0	1	0	0	0
6	0	0	10	3	3	6
7	0	0	9	0	0	18

The columns "FLIC 5000", "BIC 5000", and "AIC 5000" contain the preferences of the three model selection criteria for the number of components in a normal mixture model for birthweight distribution, based on 25 samples of size 5000 from the population of white singletons born to heavily smoking mothers. The next nine columns correspond to sample sizes of 10000, 25000, and 50000.

ISSN: 1471-2393