5033685049

Build Type

github

Committed by GitHub

Commit Message

Merge 938d524d8 into 7ddf8cae8

Pull Request Pull Request #78: Add f-score and custom analysis module

Run Details

126 of 157 branches covered (80.25%)

Branch coverage included in aggregate %.

55 of 71 new or added lines in 7 files covered. (77.46%)

40 existing lines in 4 files now uncovered.

377 of 437 relevant lines covered (86.27%)

1.73 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

64.0

/src/stocktracer/analysis/f_score.py

"""Piotroski F-score is a number between 0 and 9 which is used to assess strength of company's financial position."""
import logging
from typing import Optional

import pandas as pd
from beartype import beartype

from stocktracer.analysis.annual_reports import create_normalized_sec_table
from stocktracer.data.sec import Filter as SecFilter
from stocktracer.interface import Analysis as AnalysisInterface

logger = logging.getLogger(__name__)


@beartype
class Analysis(AnalysisInterface):
    """

    The score is used by financial investors in order to find the best value
    stocks (nine being the best). The score is named after Stanford accounting professor Joseph Piotroski.

    Calculation procedure
    =====================

    The score is calculated based on 9 criteria divided into 3 groups.[2]

    - Profitability
        - Return on Assets (ROA) (1 point if it is positive in the current year, 0 otherwise);
        - Operating Cash Flow (1 point if it is positive in the current year, 0 otherwise);
        - Change in Return of Assets (ROA) (1 point if ROA is higher in the current year compared to the previous one, 0 otherwise);
        - Accruals (1 point if Operating Cash Flow/Total Assets is higher than ROA in the current year, 0 otherwise);
    - Leverage, Liquidity and Source of Funds
        - Change in Leverage (long-term) ratio (1 point if the ratio is lower this year compared to the previous one, 0 otherwise);
        - Change in Current ratio (1 point if it is higher in the current year compared to the previous one, 0 otherwise);
        - Change in the number of shares (1 point if no new shares were issued during the last year);
    - Operating Efficiency
        - Change in Gross Margin (1 point if it is higher in the current year compared to the previous one, 0 otherwise);
        - Change in Asset Turnover ratio (1 point if it is higher in the current year compared to the previous one, 0 otherwise);

    Some adjustments that were done in calculation of the required financial ratios are discussed in the original paper.[2]

    The score is calculated based on the data from financial statement of a company. A company gets 1 point for each met criterion. Summing up of all achieved points gives Piotroski F-score (number between 0 and 9).
    """

    under_development = True
    years_of_analysis = 2

    def analyze(self) -> Optional[pd.DataFrame]:
        # Create the filter we'll use to scrape the results
        sec_filter = SecFilter(
            # tags=["EarningsPerShareDiluted"],
            years=self.years_of_analysis,
            last_report=self.options.last_report,
            only_annual=True,  # We only want the 10-K
        )

        table = create_normalized_sec_table(sec_filter, self.options)

        assert (
            self.options.last_report.year + 1
            not in table.data.index.get_level_values(1)
        )

        logger.debug(f"filtered_data:\n{table.data}")

        f_score = pd.DataFrame(
            table.data["OperatingIncomeLoss"].div(table.data["Assets"]), columns=["ROA"]
        )
        f_score = f_score.join(table.data["OperatingIncomeLoss"])
        f_score["delta-ROA"] = f_score.groupby(by=["ticker"]).diff()[
            "OperatingIncomeLoss"
        ]
        # f_score.pivot_table(index=['ticker'], columns=['fy','OperatingIncomeLoss'])
        logger.debug(f"f_score:\n{f_score}")
        # assert f_score.loc["AAPL"].loc[2022]["ROA"] == 0.279144
        # assert f_score.loc["AAPL"].loc[2022]["OperatingIncomeLoss"] == 9.822467e10
        # assert f_score.loc["AAPL"].loc[2022]["delta-ROA"] == 1.850233e10
        return f_score

    # Reuse documentation from parent
    analyze.__doc__ = AnalysisInterface.analyze.__doc__

1	"""Piotroski F-score is a number between 0 and 9 which is used to assess strength of company's financial position."""
2	import logging	2✔
3	from typing import Optional	2✔
4
5	import pandas as pd	2✔
6	from beartype import beartype	2✔
7
8	from stocktracer.analysis.annual_reports import create_normalized_sec_table	2✔
9	from stocktracer.data.sec import Filter as SecFilter	2✔
10	from stocktracer.interface import Analysis as AnalysisInterface	2✔
11
12	logger = logging.getLogger(__name__)	2✔
13
14
15	@beartype	2✔
16	class Analysis(AnalysisInterface):	2✔
17	"""
18
19	The score is used by financial investors in order to find the best value
20	stocks (nine being the best). The score is named after Stanford accounting professor Joseph Piotroski.
21
22	Calculation procedure
23	=====================
24
25	The score is calculated based on 9 criteria divided into 3 groups.[2]
26
27	- Profitability
28	- Return on Assets (ROA) (1 point if it is positive in the current year, 0 otherwise);
29	- Operating Cash Flow (1 point if it is positive in the current year, 0 otherwise);
30	- Change in Return of Assets (ROA) (1 point if ROA is higher in the current year compared to the previous one, 0 otherwise);
31	- Accruals (1 point if Operating Cash Flow/Total Assets is higher than ROA in the current year, 0 otherwise);
32	- Leverage, Liquidity and Source of Funds
33	- Change in Leverage (long-term) ratio (1 point if the ratio is lower this year compared to the previous one, 0 otherwise);
34	- Change in Current ratio (1 point if it is higher in the current year compared to the previous one, 0 otherwise);
35	- Change in the number of shares (1 point if no new shares were issued during the last year);
36	- Operating Efficiency
37	- Change in Gross Margin (1 point if it is higher in the current year compared to the previous one, 0 otherwise);
38	- Change in Asset Turnover ratio (1 point if it is higher in the current year compared to the previous one, 0 otherwise);
39
40	Some adjustments that were done in calculation of the required financial ratios are discussed in the original paper.[2]
41
42	The score is calculated based on the data from financial statement of a company. A company gets 1 point for each met criterion. Summing up of all achieved points gives Piotroski F-score (number between 0 and 9).
43	"""
44
45	under_development = True	2✔
46	years_of_analysis = 2	2✔
47
48	def analyze(self) -> Optional[pd.DataFrame]:	2✔
49	# Create the filter we'll use to scrape the results
NEW 50	sec_filter = SecFilter(	×
51	# tags=["EarningsPerShareDiluted"],
52	years=self.years_of_analysis,
53	last_report=self.options.last_report,
54	only_annual=True, # We only want the 10-K
55	)
56
NEW 57	table = create_normalized_sec_table(sec_filter, self.options)	×
58
NEW 59	assert (	×
60	self.options.last_report.year + 1
61	not in table.data.index.get_level_values(1)
62	)
63
NEW 64	logger.debug(f"filtered_data:\n{table.data}")	×
65
NEW 66	f_score = pd.DataFrame(	×
67	table.data["OperatingIncomeLoss"].div(table.data["Assets"]), columns=["ROA"]
68	)
NEW 69	f_score = f_score.join(table.data["OperatingIncomeLoss"])	×
NEW 70	f_score["delta-ROA"] = f_score.groupby(by=["ticker"]).diff()[	×
71	"OperatingIncomeLoss"
72	]
73	# f_score.pivot_table(index=['ticker'], columns=['fy','OperatingIncomeLoss'])
NEW 74	logger.debug(f"f_score:\n{f_score}")	×
75	# assert f_score.loc["AAPL"].loc[2022]["ROA"] == 0.279144
76	# assert f_score.loc["AAPL"].loc[2022]["OperatingIncomeLoss"] == 9.822467e10
77	# assert f_score.loc["AAPL"].loc[2022]["delta-ROA"] == 1.850233e10
NEW 78	return f_score	×
79
80	# Reuse documentation from parent
81	analyze.__doc__ = AnalysisInterface.analyze.__doc__	2✔

gyund / fundamental-analysis / 5033685049

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous