How to estimate theta? about girth HOT 4 CLOSED

Ability estimates are scheduled for the April release, I will try to hasten it if it's a pressing concern?

from girth.

The branch "abilities_estimation" has included maximum a posteriori (MAP) and maximum likelihood (MLE). Going to add Expected a posterior (EAP) and unittests next.

from girth.

Added Expected a posteriori, need to write unittests before committing.

from girth.

Committed this into master, add three ability estimation methods:

• Maximum Likelihood
• Maximum a Posteriori
• Expected a Posteriori

import numpy as np
from girth import twopl_separate
from girth import create_synthetic_irt_dichotomous
from girth import ability_eap

# Create Synthetic Data
difficulty = np.linspace(-2, 2, 10)
discrimination = 0.5 + np.random.rand(10) * 2
theta = np.random.randn(400)

syn_data = create_synthetic_irt_dichotomous(difficulty, discrimination, theta)

alpha, beta = twopl_separate(syn_data)
theta_star = ability_eap(syn_data, beta, alpha)

list(zip(theta.round(3), theta_star.round(3)))

from girth.