is it not still arbitrage? there should be no difference in returns, when comparing an investment based on the derivative, and an investment based on the original instrument/ security

Note: This is related to your other question, I am unable to find that.

Dear researcher, firstly, when you have used robust methodologies such as 2SLS and GMM, you do not need to bother about diagnostics you are discussing. Secondly, these models potentially deal with endogeneity, and I have not seen the diagnostic test in papers accompanying these models. Thirdly, still if diagnostic is important for you, you check Heteroskedasticity and serial correlation. If your model is free from both no problem if residuals are normally distributed or not. You can continue with the same model.

I suggest you, don’t bother to report these diagnostics with the results of GMM or 2SLS. If you can follow the pattern of some good papers for reporting the required results, it would be beneficial for you.

All the best for you research.

thank you prof. Muhammad Asif Khan for your valuable answer

Dear Srikanth

You can employ statistical techniques to discover where price discovery occurs, how it spreads to other markets, and what (e.g. news) caused it to happen. See the following literature for specific methods and empirical support demonstrating price discovery that can be applied to derivatives:

Article One Security, Many Markets: Determining the Contributions to...

Article Real-Time Price Discovery in Global Stock, Bond and Foreign ...

https://patents.google.com/patent/US5845266A/en

http://www.aaajournals.org/doi/abs/10.2308/accr.2003.78.1.193?code=aaan-site

Article Analysis of stock market manipulations using knowledge disco...

https://academic.oup.com/rfs/article/27/8/2267/1582754

Article Price Discovery in American and British Property Markets