Table 2.
Model Fit Estimates and Variance Parameters: Alternative Models Predicting Weight, by Time to Death, of 800 Male Decedents From the Baltimore Longitudinal Study of Aging, Maryland, 1958–2005
| Model, Description | Model 1: Linear | Model 2: Quadratic | Model 3: 2-Period Linear | Model 4: 2-Period Quadratic | Model 5: 3-Period Linear |
| AIC | 36,942.4 | 36,756.0 | 36,084.9 | 35,826.1 | 35,783.0 |
| Residual (SE) | 7.61 (0.15) | 7.40 (0.15) | 6.07 (0.12) | 5.40 (0.11) | 5.36 (0.11) |
| Variance in model estimates (SE) | |||||
| A: Intercept | 180.88 (9.89) | 177.44 (9.68) | 181.75 (10.65) | 174.51 (10.95) | 172.56 (10.80) |
| B: Time to death | 0.19 (0.01) | 0.17 (0.01) | 0.38 (0.03) | 0.63 (0.06) | 0.63 (0.06) |
| C: Time >k1 | 0.54 (0.05) | 1.01 (0.12) | 0.98 (0.11) | ||
| D: (Time >k1)2 | 0.0006 (0.00008) | ||||
| E: Time >k2 | 0.57 (0.07) | ||||
| Covariance of A and B | −3.37 (0.31) | −3.11 (0.29) | −4.23 (0.46) | −4.14 (0.65) | −4.10 (0.65) |
| Covariance of A and C | 1.72 (0.55) | 0.15 (0.82) | 0.56 (0.80) | ||
| Covariance of A and D | 0.07 (0.02) | ||||
| Covariance of A and E | 1.53 (0.62) | ||||
| Covariance of B and C | −0.38 (0.04) | −0.57 (0.08) | −0.59 (0.07) | ||
| Covariance of B and D | −0.004 (0.002) | ||||
| Covariance of B and E | −0.07 (0.05) | ||||
| Covariance of C and D | −0.01 (0.00008) | ||||
| Covariance of C and E | −0.35 (0.07) |
Abbreviations: AIC, Akaike Information Criterion; SE, standard error.