Data set error: "Error using set, values must be..."

announcements (new versions, changes, bugs, installation problems...) related to M-SURGE, E-SURGE and U-CARE

Data set error: "Error using set, values must be..."

Postby Luke Wakeen » Mon Jul 16, 2012 12:00 pm

HI,

I am running E-surge for the first time and I keep running into an error. I am using the MARK input format. Although my data set is processed fine in U-Care, E-surge seems to have a problem with my data set. The data dialogue in the main E-surge window shows I have 7 events when really my data has 6 events. I have used the gui dialogue to change this but E-surge will not accept the change. Then, when I run the model it the computation process dies complaining "Error using set Values must be monotonically increasing." It then gives me an astronomical AIC.

So far I have double checked my GPAT matrices very carefully, and tried using a very simple GEMACO input. These don't appear to be the problem. I have searched my data for the number 7, which should not show up, and it does not show up. Finally, I have run a simpler version of the same data set (few states), and produced the expected result: a reasonable AIC and the full ouput--which is to say it runs fine.

One last consideration is that this model is supposed to take experience into account so the numbers 2 and 3 will appear in each history only once, if at all. Could this be causing a problem?

Thank you

Luke
Luke Wakeen
 
Posts: 3
Joined: Fri Jul 13, 2012 3:48 pm

[SOLVED]Re: Data set error: "Error using set, values must be

Postby Luke Wakeen » Tue Jul 17, 2012 11:55 am

Figured it out. It was an error in coding my data set which implied transitions that conflicted with Gepat.
Luke Wakeen
 
Posts: 3
Joined: Fri Jul 13, 2012 3:48 pm


Return to software problems/news

Who is online

Users browsing this forum: No registered users and 4 guests