Abstract

The computer program supports constructing quadratic objective functions in n≤ 6 target variables for decision making (in n ≤ 5 target variables for monotonic quadratic objective functions). The model (Tangian 2001) is based on fitting a quadratic objective function with N = (n 4 + 2)(n + l)/2 coefficients to its m ≥ N known values obtained from interviewing a decision maker. Depending on the interview scenario, the data on preferences are either ordinal, or cardinal.The user’s goal is making a document with tables and figures representing the model’s input and output. The model’s input includes the interrogation table ‘Joint questionnaire’, and the table ‘Alternatives’. The former contains information about the decision maker’s preference, the latter contains real alternatives to be evaluated with the objective function. The model’s output includes the table ‘Quadratic objective function’ with coefficients, graphs of the its indifference curves, the evaluation of the alternatives in the input tables, and some service information like fit errors. The user can either type numerical values into tables, or edit indifference curves by dragging-and-dropping their points in graphs.KeywordsEconometric decision modelquadratic objective functionquadratic utilitymonotonicityinterview designordinal stability of preferencesmarket prices

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call