Re: how to simulate the gain and offset of dynamic comparato
The comparators usually use positive feedback, and if enough time is given they take any input and generate rail-to-rail output levels. So, it is not possible to define a "gain".
In what concerns the monte-carlo simulation you must define, for every element, the statistical variation of its parameters according to the information made available by the foundry - you should check the manual of the simulator you're using to see how to do this.
Regards.