Have an idea?

Visit Sawtooth Software Feedback to share your ideas on how we can improve our products.

Back-translation of utilities to chips in CBC HB Chip Allocation


In one of the studies, we assumed the "volumetric", chip-allocation task with answers ranging 1-100. We tricked CBC HB by creating the dummy "none". Now we need to estimate chip value of individual levels of some attributes. How to back-translate the resp-level utilities into chips?

I noticed that while utilities vary from -1.6 to +2.0, immense majority lies between -1 and +1. Hence I was considering TANH (arcus tangent) function to trim down the "outliers" and enhance differences in the mainstream. The results seems to be quite satisfactory, but perhaps you can offer a smarter approach?


asked Mar 18, 2020 by Piotr (185 points)

1 Answer

0 votes
You estimate distributed chips for choice alternatives  using the MNL equation and then multiplying by the largest volume ever given for each respondent.

This means you need to assume a market simulation scenario where there are multiple alternatives that have differing attribute levels per alternative.  Plus, the None alternative needs to be one of the alternatives.  Next, use the MNL equation to calculate the probabilities for the alternatives (including the None) according to the utilities.

Next, multiply the probabilities by the largest volume for each respondent.
answered Mar 19, 2020 by Bryan Orme Platinum Sawtooth Software, Inc. (189,140 points)
Yes, I know how to create a simluator with scenarios, but that is not the goal here. We are to derive "value" (in chips) of each level of each attribute. Kindly share your POV about the algorithm described in the qn
A gentler reminder about the request to answer the above question
I'm sorry, I don't have a new insight into your question.  I've never faced that situation before.