Thank you. I tried with a Michaelis-Menten (more biologically relevant) but it does not fit:
x = collect(0:2.8:60)
y = V
par = [0.0, 1.0]
model = michaelis_menten(x, p) = (x .* p[1]) ./ (x .+ p[2])
fit = curve_fit(model, x, y, par)
P = fit.param
Y = michaelis_menten(x, P)
I am trying with a logistic but it also does not give good results:
model = logistic(x, p) = ((p[1].*x) .* (1 .- (x./p[2])))
fit = curve_fit(model, x, y, par)
With a ‘real’ logistic, I get a bit better but still a lot off:
par = [10^7, 0.0, 1.0] # par = [0.0, 1.0]
model = true_logistic(x, p) = p[1] ./ (1 .+ exp.(-p[2].*(x.-p[3])))
fit = curve_fit(model, x, y, par)
How can I improve the regression with these functions?