why simulated differ from the datasheet?

hi,if i use the correct biasing of ATF36077 at vgs=-.2,vds=1.5,id=10 m A using the nonlinear module at ADS,the simulated s par.differ very large from the datasheet ,what i do,can i use the s par.module or using the new s par. of the simulated one and using two port sumbole to define the new s par. iand using it in the design of matching?
regards

Use the S parametes of the DATA SHEET for a linear design, while the non linear model for a non linear design.

The suggestion marcomdd made is valid but no transistor is similar to the next one even if it is produced in the same line of manufacturing.

So do not expect what has been given in datasheet or simulated for that purpose ever going to match at least in high frequencies.

Of course there improvements. Look what people at Intel Corporation do with band-gap engineering.

For yourself try to give a best average might be logical.