Three-way metric unfolding via alternating weighted least squares (Q1079311): Difference between revisions

A three-way unfolding problem finds points to fit \(\delta_{ijk}\) as subjective distance between ideal point of judge j and stimulus k in situation i. The author shows that the problem of degeneracy can be reduced by using weights in the least squares, such as \(\delta^{- p}_{ijk}\) for some p, and gives some examples.