fitting

sharendr · Post by **sharendr** » Tue Jul 19, 2005 5:14 am

Hi!

When I fit the datas, I couldn't fit (0,0) point as one of my observations? How can I fit that point? The program didn't allow .

thank you

harendra

Jirka · Post by **Jirka** » Tue Jul 19, 2005 4:10 pm

harendra

I do not know if I understand correctly and which program within the STANMOD family of programs you are using. Does (0.0) mean zero time and zero spatial coordinate. Obviously it does not make much sense to fit concentration at time zero, since that does not depend on any parameters. Time zero is given as the initial condition.

J.

sharendr · Post by **sharendr** » Wed Jul 20, 2005 4:14 am

Hi!

The (0,0) indicates zero time and zero concentration. In breakthrough curves that point is the initial point in effluent. So how can I include that point?

THANK YOU

harendra

ntoride · Post by **ntoride** » Wed Jul 20, 2005 1:12 pm

harendra:

CXTFIT2 originally allowed to have (time ,conc) = (0, 0) at a certain depth at the beginning of BTC data for users like you although it does not help anything to estimate parameters. As Jirka mentioned, using data at t= 0 does not make any sense to estimate parameters because the initial condition is the condition you need to specify as similar to the boundary condition. The CDE will be solved for the initial and boundary conditions. Because of this reason, the STNAMOD interface dose not allow (0,0) in the observed data.

Nobuo