compare cagr result with another python package

[sugizo]

env
google colab

code

pip install -U QuantStats openseries okama

from openseries import OpenTimeSeries
import yfinance as yf
import quantstats as qs
import okama as ok

ticker = 'AAPL' 
period = '5y' # 1d, 5d, 1mo, 3mo, 6mo, 1y, 2y, 5y, 10y, ytd, max, etc
asset_list = ['AAPL.US']
ccy = 'USD'
first_date = '2020-05'

# openseries
stock = yf.Ticker(ticker = ticker)
history = stock.history(period = period)
series = OpenTimeSeries.from_df(dframe = history.loc[:, "Close"] )
print('Compound Annual Growth Rate (CAGR) :', series.geo_ret) 

# quantstats
qs.extend_pandas()
stock = qs.utils.download_returns(ticker = ticker, period = period)
print('Compound Annual Growth Rate (CAGR) :', stock.cagr() )

# okama
x = ok.AssetList(asset_list, ccy = ccy, first_date = first_date)
#x.describe(years = [1, 2, 3, 4, 5] )
x.get_cagr()

result

question
why the cagr result value differs is so far with another python module ?

thanks and best regards

[chilango74]

@sugizo Thank you for interesting comparission.
The main reason of the difference is that okama uses adjusted close time series. Adjusted close gives total return (adjusted for dividends). It looks like in quantstats uses simple close without adjusting.

The difference with openseries might be explained by different time frame. You use period 5y, but not necesary the first date is the same. And finally, when you get ok.AssetList(asset_list, ccy = ccy, first_date = first_date) in okama you receive the data together with inflation by default. However, there is no inflation data for May yet. So, you need to check which is the last date. Probably it is 2025-04. If you want the latest data use inflation=False: ok.AssetList(asset_list, ccy = ccy, first_date = first_date, inflation=False) .

[sugizo]

the close different result between
openseries
and
okama
when execute
ok.AssetList(asset_list, ccy = ccy, first_date = first_date, inflation = False)
which is
0.238633

while
openseries executed on first posted result is
0.22544472621954958

whole result

thanks and best regards

[sugizo]

openseries formula for cagr

ref
CaptorAB/openseries#291

p.s.
there is no judgement
for any python package

just a discussion place to learn (food for thought)

thanks and best regards

[chilango74]

The formula is the same in okama. The difference is that we use monthly data. And openseries - daily. So the data frame is not the same,