Contribution to the computation of daily stock return, i.e. a preprocessing module

[hovinh]

Hello,

I wonder if it's a good idea to extend this package to cover the data preprocessing, i.e. transform data into ready-to-consumed before feeding it into this package.

To elaborate further, when I research event study, I realize there are two different ways to compute daily stock return, that are:

• $R_t = St/S{t-1} -1$ with $S_t$ is stock value at day $t$ and $R_t$ is the correspondingly computed day-to-day return.
• $R_t = log(St/S{t-1})$ similarly but have good distributional characteristics.

Also, when I work on a use case where data could span in both negative and positive zones, I must develop a new variant for the first approach to handle corner cases dealing with 0. I'm new to this and am unsure if an event study can be applied to this case.

Please let me know if this idea is good and if I can contribute to this feature. Thank you :))

[hovinh]

I just realized this is documented in the API. Still, if the case of positive + negative can be an extension, I'm happy to help :)) Or, if the data is missing some of the date, we can populate them with value from the day before.

https://lemairejean-baptiste.github.io/eventstudy/api/eventstudy.Single.import_returns.html#eventstudy.Single.import_returns

[bkrayfield]

@hovinh That's a great idea. I think there should be more messaging around missing data, or missing tickers. Do you want to make a branch?

[hovinh]

Sure, I could start working on it next week and update this thread. I will do a PR once it is ready.

[hovinh]

Hi @bkrayfield , after reviewing the API, I would like to clarify the contribution:

• Objective: Add more messages around missing data or missing tickers.
• Observation: Single.__init__() is the "entry" function for model_data to be processed by any model of choice, hence an additional Single._assert_data_quality() to be invoked within Single.__init__() can be used for data check. This function can check for data availability and output warning message based on given parameters model_data, event_date, event_window, estimation_size, buffer_size.
• Implementation:
  • Single._assert_data_quality()
  • utils.Logger: a class to handle logging.

Could you help confirm my approach and let me know any additional implementation?