For the ESD I suspect 2kV Human Body Model. Most of our devices conform to that unless it connects to the outside world - like with an RS232 chip in which case it will be higher .
Regarding the loop response, this depends on the transformer and the output cap and its ESR, so this is why we would not specify it - it is determined too much by the external components that we have no control over. Indeed I don't think we specify the loop characteristics of any of our dc/dc converters for this reason. If you really need to work out the loop response, you need to apply a step load to the output and see how it rings. If it rings excessively, increase the output capacitance (since you cannot compensate the output of the error amplifier). If you really want optimum loop response, the LT3748 is a flyback that gives you access to the output of the error amplifier so you can tailor the loop response to your needs. In this case, you will need a Bode Analyser to inject a signal in to the loop to measure the gain and phase margin. I recommend the Omicron Bode 100 from personal experience.
You could try LTspice, but the LTspice models are designed for transient analysis (time on the x axis) not frequency analysis (freq on the x axis) so the models probably will not give you a true representation of the loop characteristics