[R] uniroot

dpender d.pender.1 at research.gla.ac.uk
Fri Feb 4 13:35:00 CET 2011


Hi,

I am using the uniroot function in order to carry out a bivariate Monte
Carlo simulation using the logistics model.

I have defined the function as:

BV.FV <- function(x,y,a,A)
(((x^(-a^-1)+y^(-a^-1))^(a-1))*(y^(a-1/a))*(exp(-((1^(-a^-1)+y^(-a^-1))^a)+y^-1)))-A

and the procedure is as follows:

Randomly generate values of A~(0,1), y0 = -(lnA)^-1

Where: A=Pr{X<xi|Y=yi-1} and a is the dependency between X and y (0.703)

Use y0 to determine x where x = x(i) and y = y0(i-1) in the above equation.

Unfortunately when the randomly defined A gets to approximately 0.46 the
intervals I provide no longer have the opposite signs (both -ve).

I have tried various different upper limits but I still can't seem to find a
root.

Does anyone have any suggestions?

Thanks,

Doug

-- 
View this message in context: http://r.789695.n4.nabble.com/uniroot-tp3260090p3260090.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list