Constrained least-squares estimation in Python

python optimization scipy least-squares
Nick picture Nick · Feb 14, 2012 · Viewed 18k times · Source

I'm trying to perform a constrained least-squares estimation using Scipy such that all of the coefficients are in the range (0,1) and sum to 1 (this functionality is implemented in Matlab's LSQLIN function).

Does anybody have tips for setting up this calculation using Python/Scipy. I believe I should be using scipy.optimize.fmin_slsqp(), but am not entirely sure what parameters I should be passing to it.[1]

Many thanks for the help, Nick

[1] The one example in the documentation for fmin_slsqp is a bit difficult for me to parse without the referenced text -- and I'm new to using Scipy.

Answer

denis picture denis · Mar 31, 2012

scipy-optimize-leastsq-with-bound-constraints on SO givesleastsq_bounds, which is leastsq with bound constraints such as 0 <= x_i <= 1. The constraint that they sum to 1 can be added in the same way.
(I've found leastsq_bounds / MINPACK to be good on synthetic test functions in 5d, 10d, 20d; how many variables do you have ?)

Related questions

scipy.optimize.leastsq with bound constraints

I am looking for an optimisation routine within scipy/numpy which could solve a non-linear least-squares type problem (e.g., fitting a parametric function to a large dataset) but including bounds and constraints (e.g. minima and maxima for the …

Read More
how to find global minimum in python optimization with bounds?

I have a Python function with 64 variables, and I tried to optimise it using L-BFGS-B method in the minimise function, however this method have quite a strong dependence on the initial guess, and failed to find the global minimum. But …

Read More
Restrict scipy.optimize.minimize to integer values

I'm using scipy.optimize.minimize to optimize a real-world problem for which the answers can only be integers. My current code looks like this: from scipy.optimize import minimize def f(x): return (481.79/(5+x[0]))+(412.04/(4+x[1]))+(365.54/(3+x[2]))+(375.88/(3+x[3]))+(379.75/(3+x[4]))+(632.92/(5+x[5]))+(127.89/(1+x[6]))+(835.71/(6+…

Read More