Adjust for stock splits in portfolio engine #4195

rajanprabu · 2023-02-09T23:18:39Z

As we discussed over twitter ( https://twitter.com/jrp980/status/1623386325142347784?s=46&t=_PF5iVHF-dn4emakfIM6ag) about the stock split bug in portfolio_engine. Here is the fix to adjust for stock splits.

#4182

here is the PR for that. Its done in step 5 of __preprocess_transactions.

jmaslek · 2023-02-10T02:46:22Z

openbb_terminal/portfolio/portfolio_engine.py

+                    if not splits_df.empty :
+                        splits_df = splits_df.tz_localize(tz=None)
+                        for split_date in splits_df.index :
+                            self.__transactions['Quantity'] = np.where((self.__transactions['Ticker'] == ticker) & (self.__transactions['Date'] < split_date), self.__transactions['Quantity'] * splits_df.loc[split_date].values , self.__transactions['Quantity'])


I think you have the condition reversed. If the date is > split_date, then we multiply. For np.where it yields the first value if condition is True.

We also need to be careful and make sure that get_splits is sliced so that anything that happened before our initial purchase doesn't count )i.e, splits_df = splits_df[splits_df.index > self.__transactions['Date'][0]]

If we are holding the scrip before the event (split) then quantity will be multiplied. I think that the condition is right. I ran with Apple as an example and worked fine. For example AAPL 1 share bought before first split date 1987-06-16 should be 224 shares now.

Slicing is also taken cared off as it iterates through splits. AAPL shares bought during 2010 should be 28 shared today. To be on the safe maybe side split_df can be sorted using index ( date).

So I have an example. If I bought 1 share of appl at the beginning of 2005, it underwent splits of 2,7,4, so 1 share then = 56 now. Using your code says we initially bought 56:

If I switch the < to > but dont slice split, it accounts for everything historical, saying I started at 4 shares

If I slice splits to not include before we bought, we get the desired behavior:

I think I see where I am looking differently. Your method is correct if I enter into the template how many shares I bought from adjusted price. So if I bought 400$ of appl a few years ago, I only got 1 share. If you enter that as 4 shares at 100$ into your excel, then your split adjustment is correct. If you enter into the excel as 1 share at 400, then you need to do the reverse way I have here.

In the PR I have inserted the code in the portfolio_engine.py under the preprocess_transactions.. I modify the quantity immediately after reading the transactions so that rest is taken cared of. I checked with GOOG and AAPL and it worked correctly.

This was my input

Date Ticker Type Price Quantity Side Currency Fees yf_Ticker

02/11/22 GOOGL Stock 89.25 5 Buy USD 1 GOOGL

27/09/22 GOOGL Stock 99.75 5 Buy USD 1 GOOGL

26/04/22 GOOGL Stock 2400 1 Buy USD 1 GOOGL

22/04/22 GOOGL Stock 2400 1 Buy USD 1 GOOGL

from openbb_terminal.sdk import openbb transactions_path = '/Users/prabu/OpenBBUserData/portfolio/holdings/test_split.xlsx' P = openbb.portfolio.load(transactions_path) P.get_transactions()

This is the output I got..

Date Type Ticker Side Price Quantity Fees Investment Currency Sector Industry Country Region

0 02/11/22 STOCK GOOGL Buy 89.25 5 1 447.25 USD Communication Services Internet Content & Information United States North America

1 27/09/22 STOCK GOOGL Buy 99.75 5 1 499.75 USD Communication Services Internet Content & Information United States North America

2 26/04/22 STOCK GOOGL Buy 120 20 1 2401 USD Communication Services Internet Content & Information United States North America

3 22/04/22 STOCK GOOGL Buy 120 20 1 2401 USD Communication Services Internet Content & Information United States North America

I see whats conflicting here. You are applying to the data frame that contains historical price data. Im applying the logic to transactions itself ( before it gets the historical price). I found it easier to adjust price and quantity in transactions itself.

jmaslek · 2023-02-10T02:48:23Z

openbb_terminal/portfolio/portfolio_engine.py

+                        splits_df = splits_df.tz_localize(tz=None)
+                        for split_date in splits_df.index :
+                            self.__transactions['Quantity'] = np.where((self.__transactions['Ticker'] == ticker) & (self.__transactions['Date'] < split_date), self.__transactions['Quantity'] * splits_df.loc[split_date].values , self.__transactions['Quantity'])
+                            self.__transactions['Price'] = np.where((self.__transactions['Ticker'] == ticker) & (self.__transactions['Date'] < split_date), self.__transactions['Price'] / splits_df.loc[split_date].values , self.__transactions['Price'])


Also isn't this not needed because the price is already split adjusted (which is why we need to adjust quantity in the first place)? If the value is Price * Quantity and its incorrect, then Price / Split * Quantity * Split is the same and will still be incorrect.

yes Im aware that we use Adj. Close. But if we don't adjust the price here in get_transactions one will get quantity that is after split but price that is before split. Hence its introduced here.

JerBouma · 2023-02-13T09:49:38Z

I haven't looked into the code but basically what my expectation is, is that the volume and price is corrected after the split in which the user is informed this happend. So the following (values are random). Please note the remark regarding Yahoo Finance. If this source is used, the split is factored in within all historical data. We will also need an argument in which you can disable this correction if you entered values that are split-adjusted already (which you do by default if you would make an example portfolio with Yahoo Finance data)

Alternatively, you can also adjust the historical prices of Yahoo Finance which would correctly reflect the actual historical prices. I believe this is what @jmaslek does? Why I wouldn't like it as much is because it doesn't reflect your actual current positions (given that you gained a lot more shares).

Book1.xlsx

rajanprabu · 2023-02-20T20:34:32Z

@JerBouma for the example here

#4195 (comment)

Date	Ticker	Type	Price	Quantity	Side	Currency	Fees	yf_Ticker
02/11/22	GOOGL	Stock	89.25	5	Buy	USD	1	GOOGL
27/09/22	GOOGL	Stock	99.75	5	Buy	USD	1	GOOGL
26/04/22	GOOGL	Stock	2400	1	Buy	USD	1	GOOGL
22/04/22	GOOGL	Stock	2400	1	Buy	USD	1	GOOGL

with the modified code.. I got the following as portfolio returns

JerBouma · 2023-02-21T09:01:43Z

@JerBouma for the example here

#4195 (comment)

Date Ticker Type Price Quantity Side Currency Fees yf_Ticker
02/11/22 GOOGL Stock 89.25 5 Buy USD 1 GOOGL
27/09/22 GOOGL Stock 99.75 5 Buy USD 1 GOOGL
26/04/22 GOOGL Stock 2400 1 Buy USD 1 GOOGL
22/04/22 GOOGL Stock 2400 1 Buy USD 1 GOOGL
with the modified code.. I got the following as portfolio returns

Yes, this is correct. @jmaslek Can you take it from here? This is how it should be. Note again that Yahoo Finance adjusts all prices also those before the split.

jmaslek · 2023-02-21T13:15:36Z

can merge once the linters pass

codecov · 2023-03-27T22:56:46Z

Codecov Report

❗ No coverage uploaded for pull request base (develop@1fb918f). Click here to learn what that means.
Patch has no changes to coverable lines.

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #4195   +/-   ##
==========================================
  Coverage           ?   55.99%           
==========================================
  Files              ?      591           
  Lines              ?    53860           
  Branches           ?        0           
==========================================
  Hits               ?    30157           
  Misses             ?    23703           
  Partials           ?        0

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

* Adjust for stock splits in portfolio engine * Lint * Lint2 --------- Co-authored-by: James Maslek <jmaslek11@gmail.com>

reviewpad bot added the feat XS Extra small feature label Feb 9, 2023

jmaslek reviewed Feb 10, 2023

View reviewed changes

jmaslek approved these changes Feb 21, 2023

View reviewed changes

Adjust for stock splits in portfolio engine

3473c4a

rajanprabu force-pushed the fix-stock-splits branch from a243396 to 3473c4a Compare February 22, 2023 14:52

jmaslek and others added 4 commits March 2, 2023 11:16

Merge branch 'develop' into fix-stock-splits

f2b6a3d

Merge branch 'develop' into fix-stock-splits

cc8a215

Merge branch 'develop' into fix-stock-splits

10b8777

Lint

ed56f5a

jmaslek enabled auto-merge March 27, 2023 22:38

Lint2

590f670

jmaslek force-pushed the fix-stock-splits branch from 01c4287 to 590f670 Compare March 27, 2023 22:41

Merge branch 'develop' into fix-stock-splits

2b95ba8

jmaslek added this pull request to the merge queue Mar 27, 2023

Merged via the queue into OpenBB-finance:develop with commit 55c4b54 Mar 27, 2023

jmaslek added a commit that referenced this pull request Apr 13, 2023

Adjust for stock splits in portfolio engine (#4195)

fd7c01a

* Adjust for stock splits in portfolio engine * Lint * Lint2 --------- Co-authored-by: James Maslek <jmaslek11@gmail.com>

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Adjust for stock splits in portfolio engine #4195

Adjust for stock splits in portfolio engine #4195

rajanprabu commented Feb 9, 2023

jmaslek Feb 10, 2023

rajanprabu Feb 10, 2023 •

edited

Loading

jmaslek Feb 10, 2023

jmaslek Feb 10, 2023

rajanprabu Feb 10, 2023 •

edited

Loading

rajanprabu Feb 11, 2023

jmaslek Feb 10, 2023

rajanprabu Feb 10, 2023 •

edited

Loading

JerBouma commented Feb 13, 2023 •

edited

Loading

rajanprabu commented Feb 20, 2023

JerBouma commented Feb 21, 2023

jmaslek commented Feb 21, 2023

codecov bot commented Mar 27, 2023

Adjust for stock splits in portfolio engine #4195

Adjust for stock splits in portfolio engine #4195

Conversation

rajanprabu commented Feb 9, 2023

jmaslek Feb 10, 2023

Choose a reason for hiding this comment

rajanprabu Feb 10, 2023 • edited Loading

Choose a reason for hiding this comment

jmaslek Feb 10, 2023

Choose a reason for hiding this comment

jmaslek Feb 10, 2023

Choose a reason for hiding this comment

rajanprabu Feb 10, 2023 • edited Loading

Choose a reason for hiding this comment

rajanprabu Feb 11, 2023

Choose a reason for hiding this comment

jmaslek Feb 10, 2023

Choose a reason for hiding this comment

rajanprabu Feb 10, 2023 • edited Loading

Choose a reason for hiding this comment

JerBouma commented Feb 13, 2023 • edited Loading

rajanprabu commented Feb 20, 2023

JerBouma commented Feb 21, 2023

jmaslek commented Feb 21, 2023

codecov bot commented Mar 27, 2023

Codecov Report

rajanprabu Feb 10, 2023 •

edited

Loading

rajanprabu Feb 10, 2023 •

edited

Loading

rajanprabu Feb 10, 2023 •

edited

Loading

JerBouma commented Feb 13, 2023 •

edited

Loading