Pooled within-groups correlation between discriminating variables and standardized canonical
discriminant functions. Variables Function 1 Function 2 Bamboo cover -0.66 (*) Bamboo density -0.62 (*) 0.46 Bamboo height -0.55 (*) 0.05 Slope 0.08 0.66 (*) Herb cover 0.40 -0.53 (*) Number of shrubs 0.24 -0.42 (*) (*) Largest absolute correlation between each variable and the discriminan function.
Palatal rugae in population differentiation between South and North Indians: A
discriminant function analysis.
Since the dependent variable is split in two categories, one
discriminant function is estimated with one categorical dependent and seven independent variables.
According to the equation, the
discriminant function score for any set of measurements that is above the cut score is probably for a male individual whereas a score that was below the cut score is probably for a female subject.
Table VI.- Percentage of specimens classified in each group from original and cross-validation tests of
discriminant function analysis (DFA) on truss-network data.
Discriminant function sexing of fragmentary and complete femora: standards for contemporary Croatia.
Last, "Sex determination from the foramen magnum:
discriminant function analysis in an eighteenth and nineteenth century British sample," International Journal of Legal Medicine, vol.
The results showed that two statistically significant
discriminant functions were formed: Wilks' lambda = 0.044, [X.sup.2] = 86.010, df = 16, and p value = 0.000 for the first function and Wilks' lambda = 0.308, [X.sup.2] = 32.403, df = 7, and p value = 0.000 for the second.
For training samples with capacity [n.sub.i] from totality [G.sub.i] (where [alpha] = 1,2, ..., [n.sub.i]; I = 1,2, ..., m), all training samples were successively substituted into the established
discriminant function and the corresponding criterion was used for water source recognition.
The significance test of the
discriminant function are shown in [Table 4], Wilks' ?' value was 0.530, Chi-square value was 137.535, so the discriminant result was proved to be effective.
Punniyamoorthy and Thoppan (2012a) have tested for the assumptions governing the use of the Linear
Discriminant Function. The assumptions are that the data should be normally distributed (verified using the Q-Q Plot) and that the two groups should have equal variance-covariance matrices (tested using the Box's M Test).
A linear
discriminant function, y(x), that can reduce the data to a one-dimensional numerical value is determined:
