Thinking outside the curve, part II: modeling fetal-infant mortality

Table 4 Mixture Modeling Results for White Singleton Infants

Quantity	p₁	p₂	p₃	p₄
$\overset{\land}{θ}$ [average of 25 estimates]	.005	.117	.810	.068
${\overset{\land}{S}}_{θ}$ [standard deviation of 25 estimates]	.001	.010	.012	.006
${\overset{\land}{B}}_{θ}$ [bias adjustment]	.001	.012	.011	.007
Confidence interval	(.004, .006)	(.099, .135)	(.792, .827)	(.058, .078)
Quantity	μ₁	μ₂	μ₃	μ₄
$\overset{\land}{θ}$ [average of 25 estimates]	862	2948	3402	4056
${\overset{\land}{S}}_{θ}$ [standard deviation of 25 estimates]	60	52	6	18
${\overset{\land}{B}}_{θ}$ [bias adjustment]	22	47	4	36
Confidence interval	(809, 915)	(2874, 3021)	(3395, 3410)	(4011, 4100)
Quantity	σ₁	σ₂	σ₃	σ₄
$\overset{\land}{θ}$ [average of 25 estimates]	233	776	421	416
${\overset{\land}{S}}_{θ}$ [standard deviation of 25 estimates]	40	23	5	19
${\overset{\land}{B}}_{θ}$ [bias adjustment]	42	25	5	11
Confidence interval	(170, 295)	(739, 813)	(413, 429)	(395, 437)

Parameters in a 4-component normal mixture model for birthweight distribution are estimated, based on 25 samples of size 50,000 from the population of white singletons in general. Confidence intervals are constructed using Equations (6) and (7) with C ₀ = 2.5 and φ = .0055.

ISSN: 1471-2393