This Case is aboutÂ ANALYTICS

PUBLICATION DATE: November 07, 2011 PRODUCT #: UV6335-PDF-ENG

It draws on the analogy between building and modeling distinct selection a regression model using a dummy variable that is dependent as well as on an example that exemplifies the requirement for estimating the likelihood of a selection instead of the selection itself, which results in a particular type of regression – logistic regression. The note presents a random utility selection model, of which the logistic regression model is the most frequently used and the notions of utility. It reveals how selection probabilities may be built from utilities resulting in the logit model. Working through a comprehensive example using accompanying spreadsheet model and Solver, the note gives pupils profound understanding for MLE works and the way that it’s different and similar to the conventional least-squared approximation in linear regression.

Modeling Discrete Choice Categorical Dependent Variables, Logistic Regression, and Maximum Likelihood Estimation Case SolutionThe note concludes by delivering the outcomes of approximation using a commercial statistical applications, StatTools. The note prevents using heavy mathematical machines but nevertheless needs basic understanding of exponent and logarithmic functions, probability, and optimization with Solver, along with acquaintance with the “normal” linear regression. Programs include construction of models for conjoint analysis, estimating cost elasticity, cost optimization, merchandise versioning, product line design, and consumer selection.