Problem with subfunction within lsqnonlin??

I got the following error which involves matlab not knowing how to multiply a scalar with a sparse structure (which is just another number). Does anyone know what is going on?

Error using * MTIMES is not supported for one sparse input and one single input.

Error in trdog (line 30) grad = D*g;

Error in snls (line 315) [sx,snod,qp,posdef,pcgit,Z] = trdog(x,g,A,D,delta,dv,...

Error in lsqncommon (line 175) [xC,FVAL,LAMBDA,JACOB,EXITFLAG,OUTPUT,msgData]=...

Error in lsqnonlin (line 237) [xCurrent,Resnorm,FVAL,EXITFLAG,OUTPUT,LAMBDA,JACOB] = ...

Error in DCE_Analysis (line 35) [Pfinal, chi2] = lsqnonlin(@(P)lSqErr(insignal,inserr,tarr,AIFp,R1T,P),Pstart,Plb,Pub,options);

[Becket Hui]

I think I figured it out, it appears that everything must be type double (can't be single) or else it will fail.

Answer 1

[Steve Grikschat]

All values returned to lsqnonlin must be of data type double.

[Pranav Pathak]

Hello Sir,

Do we have any work around for this case?

I need to pass a single datatype value to the lsqnonlin function.

With Typecasting, I am able to resolve the error but this procedure does NOT optimize the value. It stops after one iteration.

Please let me know your feedback.