I had to rebuilt my standart lm regression into a nls regression as I had to determine a lower bound for one of my variables:
NONLinear <- nls (PD04_AL ~ a * Health_Care + b * Utilities + c * TALOG + d * CFOP + e * GROSS_MARGIN + f * AbsPAPER1 + g * TOT_DEBT_TO_TOT_CAP + h * RETURN_ON_ASSET + i, data = full,
start = c(a =0, b= 0, c = 0, d = 0,e = 0,f = 0,g = 0, h = 0, i = 0),algorithm = "port", na.action = "na.exclude",lower = c(a = -Inf,b= -Inf, c = -0.20,d= -Inf,e= -Inf,f= -Inf,g= -Inf,h= -Inf,i= -Inf))
The function "cooks.distance(NONLinear) I usually use does not work with nls calls. Is there any way to calculate the cooks distance quickly for my nls regression?
Thanks a lot for the help!
Daniel
