OwensQ {PowerTOST} | R Documentation |
Owen's Q-function
Description
Calculates Owen’s Q function.
Usage
OwensQ(nu, t, delta, a=0, b)
Arguments
nu |
degree of Owen’s Q |
t |
parameter t |
delta |
parameter delta |
a |
lower integration limit, only a=0 implemented |
b |
upper integration limit |
Details
Uses the relationship to non-central t-distribution (see Chou)
OwensQ = pt(t, df=nu, ncp=delta) - Integal_b_Inf(Q_integrand)
The definite integral is numerically evaluated using integrate
after a variables transformation resulting in the integration range from 0 to 1
instead of the semi-infinite original range.
This may result in higher precision and better numerical stability.
The arguments to the function must be scalars. No vectors allowed.
Value
Numeric value of Owen’s Q-function at given input arguments.
Note
This function is intended for internal use in the power calculations.
But may be useful for others.
Author(s)
D. Labes
References
Owen DB. A special case of a bivariate non-central t-distribution. Biometrika. 1965;52(3/4):437–46. doi:10.2307/2333696
Chou YM. A bivariate noncentral T-distibution with applications. Commun Stat Theory Methods. 1992;21(12):3427–62. doi:10.1080/03610929208830988
See Also
Examples
# This function is mainly intended for internal use.
OwensQ(10, 2.5, 5, 0, 2)
#should give [1] 9.388137e-06
OwensQ(10, -2.5, -5, 0, 2)
#should give [1] 0.05264363