Accurately calculating Greeks for options near expiration

Question

I understand that when a vanilla European option is near expiry, the Theta calculated from BS formula is very inaccurate and almost meaningless for practical use.

However, I'm not sure if other Greeks, such as Gamma and Vega, also have the same characteristics, i.e. becoming inaccurate near expiry.

If so, is there any other model that overcomes this problem for expiring options?

Apologies that I was not from a quant/math background; I'm asking this question from a very practical trading/risk-management perspective. I'd appreciate if anyone can point a direction for me to read further into.

Edit: I'm just looking at the simplest vanilla European options for equity index, not any exotic ones.

Answer 1

Generalizing, some people who write options trading software are not aware of a few small, but important details, resulting in some pricing idiosyncrasies. That is often the case with retail trading platforms, and you often read statements like "implied vol blows up in days before expiration", "greeks become unreliable before expiration", or suggestion of ridiculous fixes, such as using settlement time (Saturday noon) as expiration time.

That is certainly an issue, but it is not a big issue if your software is written properly. Coming from options market making background, and using better software, greeks would work more or less correctly until about 10 minutes until the close, at which point we would switch to pricing things at parity, and deltas being binary, etc...