my dependent variable is continuous and not normal but I'm not sure which family to choose ... when I run the analysis, this message appears:
restart model with a smaller learning rate or smaller step size ... it is the first time that I do this analysis and any help would be very welcome. thanks
ina <- gbm.step(data=bonaci, gbm.x = 10:33, gbm.y = 9,
family = "gaussian", tree.complexity = 5,
learning.rate = 0.01, bag.fraction = 0.5)