Moving to:
https://docs.google.com/document/d/1SC_q_t8TkLdPxglb2qcpG8q4wVe0VP-XU35FTI0ICRA/edit?usp=sharing


On Sun, Sep 13, 2020 at 12:44 AM Martin Blais <[email protected]> wrote:

> Right now, Beancount does nothing about stock splits (nor does it know
> about them either).
>
> You either (a) keep the same symbol and have a discontinuity in the price
> series, or (b) switch to a new symbol and have two sets of prices. I've
> been tending toward (a) in recent years, but in the past had used (b) --
> the extra symbol bothers me. Hasn't been much of a problem so far (though I
> invest mostly in baskets so I only have relatively few of them).
>
> The price fetchers return adjusted prices for historical prices. For
> example, if you fetch a historical price for day X before a split, the
> price returned for day X will be different if you fetch it before or after
> the split (you have to be careful with that). It's not pretty and could
> lead to errors. Even with correct recording of prices on the split date it
> could be tricky, e.g., a trade that posts the day before a split and
> settles a day after it.
>



>
> I haven't been too concerned with errors around split dates before, but
> now with my recent work on computing returns, I'm having to pay a bit more
> attention to that, and as I'm adding price plots I'm finding some small
> errors in my input file if I look for info around splits.
>
> Has anyone done some thinking around how they'd like to see stock splits
> handled?
> Obviously we cannot change the numbers in the input file itself, but I'm
> curious to hear if people have expectations or some clear idea about what
> they'd like to see in a PTA system.
>
> Some quickly jotted down ideas:
> - Checking for discontinuity in the price series to highlight errors (as a
> script, probably).
> - Adding explicit directives to record split information, and requiring
> price directives on that date.
> - Internally associating version numbers with price points based on splits
> and having operations on them which take those into account.
>
> Not so sure how to go about it yet, I'm just starting to look into this.
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Beancount" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/beancount/CAK21%2BhP-Sban-rEjJC2LqtE7P-JwkB2TvJPMHuYgQFX9mhEK0g%40mail.gmail.com.

Reply via email to