Help with MultiFinancials - historical 5 year data

[mikeystown]

Hello,

I'm having difficulty getting historical data (e.g. last 5 years) using MultiFinancials

Can somebody help with what the right code should be?

Using the following code which I saw recommended, I get the error " 'function' object has no attribute 'to_dataframe' "

company = Company(ticker)
filings = company.latest("10-K", 5)
from edgar import MultiFinancials
financials = MultiFinancials(filings)
is_df = financials.income_statement.to_dataframe()

This next set of code executes but gives the wrong data (wrong row order and missing many rows vs. 'financials = company.get_financials()' request)

company = Company(ticker)
filings = company.latest("10-K",5)
from edgar import MultiFinancials
financials = MultiFinancials.extract(filings)
is_df = financials.income_statement().to_dataframe()

Much appreciated,
MT

[Lizecheng12345]

ticker = "OXY" # 修改为目标公司代码
years = 15 # 提取5年

=====================================

获取桌面路径

=====================================

desktop_path = os.path.join(
os.path.expanduser("~"),
"Desktop"
)

=====================================

获取10-K文件

=====================================

company = Company(ticker)

filings = (
company
.get_filings(form="10-K")
.filter(amendments=False)
.head(years)
)

print("正在获取年度报告...")
print(filings)

=====================================

创建XBRLS对象

=====================================

xbrls = XBRLS.from_filings(filings)

=====================================

数据整理函数

=====================================

def clean_statement(statement):

df = statement.to_dataframe()


# 找年度日期列
period_cols = [
    c for c in df.columns
    if str(c).startswith("20")
]


# 保留原始信息
keep_cols = [
    "label",
    "concept",
    "standard_concept"
] + period_cols


df = df[
    [
        c for c in keep_cols
        if c in df.columns
    ]
]


# 删除完全没有数据的项目
df = df.dropna(
    how="all",
    subset=period_cols
)


# 调整列顺序
df = df[
    [
        "label",
        "concept",
        "standard_concept"
    ]
    +
    period_cols
]


return df

=====================================

提取三大报表

=====================================

print("正在提取损益表...")

income_statement = xbrls.statements.income_statement(
max_periods=years,
view="detailed"
)

income_df = clean_statement(
income_statement
)

print("正在提取资产负债表...")

balance_sheet = xbrls.statements.balance_sheet(
max_periods=years,
view="detailed"
)

balance_df = clean_statement(
balance_sheet
)

print("正在提取现金流量表...")

cash_flow = xbrls.statements.cash_flow_statement(
max_periods=years,
view="detailed"
)

cashflow_df = clean_statement(
cash_flow
)

=====================================

导出Excel

=====================================

income_file = os.path.join(
desktop_path,
f"{ticker}损益表.xlsx"
)

balance_file = os.path.join(
desktop_path,
f"{ticker}资产负债表.xlsx"
)

cashflow_file = os.path.join(
desktop_path,
f"{ticker}现金流量表.xlsx"
)

income_df.to_excel(
income_file,
index=False
)

balance_df.to_excel(
balance_file,
index=False
)

cashflow_df.to_excel(
cashflow_file,
index=False
)

=====================================

完成提示

=====================================

print("\n==============================")
print("数据提取完成！")
print("==============================")

print(
f"损益表文件:\n{income_file}"
)

print(
f"\n资产负债表文件:\n{balance_file}"
)

print(
f"\n现金流量表文件:\n{cashflow_file}"
) When using the code above to generate a large number of Excel financial reports for Occidental Petroleum, I encountered the issue you described. My workaround was to copy the retrieved data into ChatGPT, providing it with the original account item sequence from the XBRL source so it could reorder the data and generate a new Excel file; ChatGPT handled this task very well. However, the real problem I faced was a significant discrepancy between the large volume of data retrieved via the code and the original XBRL data. I spent a long time trying to resolve this without success. I suspect the issue arises because the original XBRL tags are inconsistent when our program merges multiple XBRL files into a single one. I haven't been able to fix this yet. I’d like to ask: is there a way to resolve this issue?