Matskási István (szerk.): A Magyar Természettudományi Múzeum évkönyve 94. (Budapest 2002)

Radoinova, D., Tenekedjiev, K. ; Yordanov, Y.: Stature estimation according to bone length in Hungarian population

IIA. Maximal stature regression together for humerus and tibia - Four outliers are rejected - 3 in the first and 1 in the second loop (n = 186-4 = 182). The GLEJSER test has detected heteroskedasticity (P-value = 0.0122). All the four models of the residuals' module are inadequate (ANOVA P-value > 0.468). This showed practically negligible heteroskedasticity. Regression formula for both bones is: (9) E[stature] = 92.62 + 0.862*H + 1.377*T (1.071) (0.175) (0.155) The correlation coefficients between the parameters of (9) are as follows: between the intercept and the humerus coefficient: -0.298, between the intercept and the tibia coefficient: 0.123 and between the humerus coefficient and tibia coefficient: -0.983. The estimated standard error is 1.03 cm with a 0.89 cm to 1.20 cm 95% confidence interval. The adjusted coefficient of multiple determination is R

2

= 0.970. The 95% confidence intervals of the model coefficients are respectively: from 89.57 cm to 95.66 cm, from 0.365 cm to 1.359 cm and from 0.936 cm to 1.819 cm. The regression parameters are significant (t-tests with P-value'

s < 0.0005) and the model (9) is adequate (ANOVA with P-value < 0.0005). The TROTTERGLESER formula did not describe the sample well (P-value = 0.000%). Fig. 9. Age correction of maximal stature according to GILES and BORCAN for males (solid line) and females (dashed line)