Consistent argument names for dividends across option pricing models

[alembcke]

Argument names for dividends across different option pricing models appear to be different, with no obvious reason for the difference. For example, tff.black_scholes.option_price takes in continuous_dividends, while tff.black_scholes.option_price_binomial takes in dividend_rates.

If there is no reason for the difference, then I would suggest changing dividend_rates to continuous_dividends, as tff.black_scholes.approximations.adesi_whaley also takes continuous_dividends. This would make it easier to write reusable code independent of the pricing model used.

What do others think of this? If everyone agrees to this change, then I can make it and submit the pull request.

[cyrilchim]

Hi Alex,

Thanks for noticing that. This is indeed an oversight. Would you like to send a PR ensuring that dividend_rates is used across the library instead of continuous_dividends. This is more aligned with discount_rates. For discrete dividends we would use something like dividend_dates and dividends.

A gracious way to rename the parameter is to use a deprecation warning for continuous_dividends, e.g., see here

Here is a simple concrete example:

from tensorflow.python.util import deprecation 

@deprecation.deprecated_args(None, "z is deprecated. Use y instead", "z")
def fn(x, y=None, z=None):
  p = deprecation.deprecated_argument_lookup("y", y, "z", z)
  return x + p

User can either use y or z but a deprecation warning is thrown for z. In the future version we remove z

Please let me know if you would like to send a PR which would be very much appreciated.

Thanks!
Cyril

[alembcke]

Hi Cyril,

Sure, I will work on this PR. I agree with using dividend_rates across the option pricing models. I also agree that including a deprecation decorator is the right way to go about such a change, so will include that in my PR. I will let you know if I have any questions.

Regards,
Alex

[cyrilchim]

Thank you Alex! This is now merged.