Dear list, I have a problem on my research project. I have three time series data, u1, u2 and v, I get the conditional distribution F1(u2|u1)and F2(v|u1) first, then produce the quantile curve by using formula: y = qgev(pnorm(rho*qnormF1(u2|u1)+sqrt(1-rho^2)*qnorm(P))),xi,mu,sigma) P,here is the different quantile: 0.05,0.1,0.5,0.9,0.95. I wonder how could I produce 3-dimentional graph? I have tried scatterplot3d(u1,u2,y), any other method? Appreciate for any reply. Thanks Mc
3-D graphing in quantile curve
2 messages · Xiaochen Sun, Rob Hyndman
One solution is to use coplot() which is a wonderful, but underused, function in R/S+. Another possibility is try the function plot.cde() in the hdrcde package. This is designed for plotting conditional density estimates. But it could probably be fooled into plotting conditional quantile estimates without too much difficulty. It will handle one or two conditioning variables. The plots are either stacked densities or highest density region strips. Or you could try persp(), image(), contour(), etc. Best wishes, Rob
Xiaochen Sun wrote:
Dear list, I have a problem on my research project. I have three time series data, u1, u2 and v, I get the conditional distribution F1(u2|u1)and F2(v|u1) first, then produce the quantile curve by using formula: y = qgev(pnorm(rho*qnormF1(u2|u1)+sqrt(1-rho^2)*qnorm(P))),xi,mu,sigma) P,here is the different quantile: 0.05,0.1,0.5,0.9,0.95. I wonder how could I produce 3-dimentional graph? I have tried scatterplot3d(u1,u2,y), any other method? Appreciate for any reply. Thanks Mc
_______________________________________________ R-SIG-Finance at stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-finance
__________________________________________________ Professor Rob J Hyndman Department of Econometrics & Business Statistics, Monash University, VIC 3800, Australia http://www.robhyndman.info/