search

lorismichel / quantregForest

R package - Quantile Regression Forests, a tree-based ensemble method for estimation of conditional quantiles (Meinshausen, 2006).
23 stars 6 forks source link

90 % prediction intervals on out-of-bag data #9

Open oziashub opened 6 years ago

oziashub commented 6 years ago

Hi. I have carried out a modelling using the function quantregForest 1.3-7. e.g. qrfozone <- quantregForest(xozone,yozone) Using plot(qrfozone,all=TRUE) one is supposed to get Estimated 90 % prediction intervals for the Ozone dataset (According to Schiesser, 2015). However, you only get the error as a function of the trees. Is there anything I am missing?

References Schiesser, Lukas. Quantile regression forests-an r-vignette. Technical report, 2015.

himal2007 commented 4 years ago

I had the same issue. I couldn't run varImpPlot.qrf function either. It would be great if someone could help us with the problem.