Rafael Arutjunjan

> The second fit gives slightly different values of parameters, but close to the right ones. > > Still not clear to me how to fake the llh fit. If...

> Hm, does not seem to use y vector. Could you confirm that it is doing exactly what we need for the LLH? Yes, it's correct. The data (i.e. xdata,...

> Let's check if we are talking about different things. > With your last passage on _exactly equivalent_, it seems to me that you have in mind **binned** likelihood fit...

> ```julia > llh(mu, sigma) = -sum(x->(x-mu)^2/sigma^2, sample) + length(sample)*log(sigma) > ``` > > is not equivalent to minimizing > > ```julia > lsq(mu, sigma) = @. (model(bins, mu, sigma)...

> We want to maintain the invariance `g(::T)::T` for out of place functions. If you are using `view`, I think you want to use the inplace version of the built...