In simbiology, getting error - "Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND = 2.971666e-16"

1 次查看（过去 30 天）

Praveen Kumar M 2019-6-17

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/467381-in-simbiology-getting-error-matrix-is-close-to-singular-or-badly-scaled-results-may-be-inaccura

关闭： MATLAB Answer Bot 2021-8-20

I am using simbiology for pharmacokinetics modelling. When i tried to use the non-mixed effect model using isqnonlin estimation method for fitting the data (in a two compartmental infusion model). I got the following error during fitting,

"Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND = 2.971666e-16"

I came across similar question being asked in context of other fields of MALTAB.

But for Simbiology, if this error occurs, how to tackle it?

2 个评论
显示无隐藏无

Sulaymon Eshkabilov 2019-6-17

Could you please be a bit more explicit and post some more details of your problem content?

Praveen Kumar M 2019-6-17

Thanks. Sorry for not being explicit. I had a pharmacokientics data of 50 patients which i tried to fit in two compartment model (drug administration type is infusion) using non-mixed effect model using isqnonlin estimation method. The error model which I kept was constant.

With this setting, I got a error that "Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND = 2.971666e-16"

But I have now found the tentative solution for this. If I change the error model to either proportional, combined or exponential, the error has gone. Now i am getting the estimated parameter values.

Regards.

请先登录，再回答此问题。