Have an idea?

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

None Utility too low

Hi Guys,

I have recently ran an acbc study where my none-utility is very very low, which is raising red flags for me.

Pretty much every simulation I run has 100% of my products being chosen.

Even when I run the calibrated none set to a purchase intent of 5/5 my none barely increases - if at all.

Any price which is mildly below average has the ability to make a product selected over and above none.

It appears that this only occurs when I use the zero-centered diffs and not the raw utilities.

So two questions -

1) is this something that sounds familiar? That has a particular fix?
2) can I just manually increase the level of the none-utility to a level where current retention is reached and then just test my products from there?

Thanks for any help,

Dean
asked Jul 7, 2014 by djtindall Bronze (600 points)
edited Jul 7, 2014 by Walter Williams
Hi Guys,

This issue only happens when I used the zero centered diffs, it is fine when I use the raw utilities.
Zero-centered diffs should not  be used in market simulations...only raw utilities.  That's probably your issue.  

Zero-centered diffs are for cross-tabulating and showing average utilities for the sample or for subgroups of the sample.  Zero-centered diffs normalizes the utilities to have (roughly) equal scale per person for comparisons.

Your solution to the original question

Please only use this to answer the original question. Otherwise please use comments.
Your name to display (optional):
Privacy: Your email address will only be used for sending these notifications.
Anti-spam verification:

To avoid this verification in future, please log in or register.
...