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

The corrélation coefficient between the regression parameters is -0.997. The resulting standard error is 0.87 cm within a 0.71 cm to 1.12 cm 95% confidence interval. The adjusted coefficient of multiple determination is R

2 = 0.972. The 95% confidence intervals of the model coefficients are from 81.29 cm to 90.01 cm and from 2.086 cm to 2.332 cm. The regression parameters are significant (t-tests with P-value'

s < 0.0005) and the model (2) is adequate (ANOVA with P-value < 0.0005). The TROTTER-GLESER formula does not describe adequately the observed sample (P-value = 0.000%). The nomogram on fibula is as shown on Fig. 2. 1.3. Maximal stature regression on tibia - Two outliers are rejected in the first loop (n = 83-2 = 81). Heteroskedasticity is detected in 7 tests: White tests (linear and squared) with P-value<0.0294, GOLDFELD-QUANDT test with P-value = 0.0146 and in 4 tests of BREUSCH-PAGAN with P-value>0.0194. Just the square model for residuals' module is adequate (ANOVA P-value = 0.0621) but is not acceptable ( = 0.045). This proved practically negligible heteroskedasticity. The regression formula for tibia is derived: (3) Efstature] = 94.22 + 1.973*T (1.85) (0.053) 185 length of fibula [cm] Fig. 2. Nomogram for predicting the maximal stature of Hungarian females using the length of fibula with age correction according to GILES and BORCAN (bold line). Three confidence margins are plotted (95% solid line, 99% dashed line, 99.9% dashdotted). The outliers are plotted with circles, and the experimental data is shown with dots