Rasch Model Estimation - Joint Maximum Likelihood

Joint Maximum Likelihood

Let denote the observed response for person n on item i. The probability of the observed data matrix, which is the product of the probabilities of the individual responses, is given by the likelihood function

\Lambda = \frac{\prod_{n} \prod_{i} \exp(x_{ni}(\beta_n-\delta_i))}{\prod_{n} \prod_{i}(1+\exp(\beta_n-\delta_i))}.

The log-likelihood function is then

\log \Lambda = \sum_n^N \beta_n r_n - \sum_i^I \delta_i s_i - \sum_n^N \sum_i^I \log(1+\exp(\beta_n-\delta_i))

where is the total raw score for person n, is the total raw score for item i, N is the total number of persons and I is the total number of items.

Solution equations are obtained by taking partial derivatives with respect to and and setting the result equal to 0. The JML solution equations are:

s_i = \sum_n^N p_{ni},\quad i=1,\dots,I
r_n = \sum_i^I p_{ni},\quad n=1,\dots,N

where . A more accurate estimate of each is obtained by multiplying the estimates by .

Read more about this topic:  Rasch Model Estimation

Famous quotes containing the words joint, maximum and/or likelihood:

    “I conjure thee, and all the oaths which I
    And thou have sworn to seal joint constancy,
    Here I unswear, and overswear them thus,
    Thou shalt not love by ways so dangerous.
    Temper, O fair Love, love’s impetuous rage,
    Be my true Mistress still, not my feign’d Page;
    I’ll go, and, by thy kind leave, leave behind
    Thee, only worthy to nurse in my mind
    Thirst to come back;”
    —John Donne (1572–1631)

    “Only at his maximum does an individual surpass all his derivative elements, and become purely himself. And most people never get there. In his own pure individuality a man surpasses his father and mother, and is utterly unknown to them.”
    —D.H. (David Herbert)

    “What likelihood is there of corrupting a man who has no ambition?”
    —Samuel Richardson (1689–1761)