Account Standardization & Time Series

An engine that normalizes raw OpenDART financial statement data into standardized accounts, producing comparable time series across companies.

Why It’s Needed

Over 2,700 Korean listed companies file financial statements in XBRL, but the same concept of “revenue” exists under different IDs and names depending on the company.

ifrs-full_Revenue          → 삼성전자
dart_OperatingIncomeLoss   → LG화학
dart_ConstructionRevenue   → 현대건설
수익(매출액)               → 카카오
영업수익                   → KB금융

DartLab unifies these into a single revenue through a 7-step mapping pipeline. Out of 15.85 million rows across 2,564 companies, 98.7% are mapped to standardized accounts.

7-Step Mapping Pipeline

Time Series Generation

Quarterly Time Series

series, periods = c.timeseries

# periods = ["2016_Q1", "2016_Q2", ..., "2024_Q4"]
# series["IS"]["revenue"]            → Quarterly revenue list
# series["BS"]["total_assets"]       → Quarterly total assets list
# series["CF"]["operating_cashflow"] → Quarterly operating cash flow list

Return structure:

series = {
    "BS": {"total_assets": [v1, v2, ...], "current_assets": [...], ...},
    "IS": {"revenue": [...], "operating_income": [...], "net_income": [...], ...},
    "CF": {"operating_cashflow": [...], "investing_cashflow": [...], ...},
}
periods = ["2016_Q1", "2016_Q2", ..., "2024_Q4"]

Periods with no data are represented as None. 0 and None have different meanings.

Normalization Method

Consolidated financial statements (CFS) are used preferentially; if unavailable, separate financial statements (OFS) are used as automatic fallback.

Direct Calls

from dartlab.engines.dart.finance import buildTimeseries, buildAnnual, buildCumulative

series, periods = buildTimeseries("005930")              # Quarterly standalone
series, years   = buildAnnual("005930")                  # Annual aggregated
series, periods = buildCumulative("005930")              # Quarterly cumulative
series, periods = buildTimeseries("005930", fsDivPref="OFS")  # Force separate statements

Value Extraction Utilities

from dartlab.engines.dart.finance import getTTM, getLatest, getAnnualValues, getRevenueGrowth3Y

series, periods = c.timeseries

# TTM (trailing 4 quarters sum) — IS, CF only
getTTM(series, "IS", "revenue")            # TTM revenue
getTTM(series, "CF", "operating_cashflow") # TTM operating CF

# Latest value — BS, IS, CF
getLatest(series, "BS", "total_assets")    # Latest total assets
getLatest(series, "IS", "net_income")      # Latest quarter net income

# Full time series list
getAnnualValues(series, "IS", "revenue")   # All-year revenue

# Revenue 3-year CAGR
getRevenueGrowth3Y(series)                 # 3-year revenue growth rate (%)

Financial Ratios (RatioResult)

r = c.ratios

# Profitability
r.roe               # ROE (%)
r.roa               # ROA (%)
r.operatingMargin   # Operating margin (%)
r.netMargin         # Net margin (%)

# Stability
r.debtRatio         # Debt ratio (%)
r.currentRatio      # Current ratio (%)
r.netDebt           # Net debt (KRW)
r.netDebtRatio      # Net debt ratio (%)
r.equityRatio       # Equity ratio (%)

# Cash Flow / Growth
r.fcf               # Free cash flow (KRW)
r.revenueGrowth3Y   # Revenue 3-year CAGR (%)

# TTM Values
r.revenueTTM            # Revenue TTM (KRW)
r.operatingIncomeTTM    # Operating income TTM (KRW)
r.netIncomeTTM          # Net income TTM (KRW)
r.operatingCashflowTTM  # Operating CF TTM (KRW)

# Latest BS Values
r.totalAssets       # Total assets (KRW)
r.totalEquity       # Equity attributable to owners of parent (KRW)
r.totalLiabilities  # Total liabilities (KRW)
r.cash              # Cash (KRW)

# Valuation (when market cap is provided)
r.per               # PER
r.pbr               # PBR
r.psr               # PSR
r.evEbitda          # EV/EBITDA

Direct call:

from dartlab.engines.dart.finance import calcRatios, buildTimeseries

series, _ = buildTimeseries("005930")
r = calcRatios(series, marketCap=400_000_000_000_000)  # Includes valuation when market cap is provided

Key snakeId Reference

BS (Balance Sheet)

IS (Income Statement)

CF (Cash Flow Statement)

The actual mapped snakeIds can number in the hundreds or thousands depending on the company. The tables above list only the most commonly used core accounts.

Cross-Company Comparison

Example comparing the revenue time series of two companies:

import dartlab

samsung = dartlab.Company("005930")
sk = dartlab.Company("000660")

s1, p1 = samsung.timeseries
s2, p2 = sk.timeseries

# Both companies accessed via the same "revenue" key
samsung_rev = s1["IS"]["revenue"]
sk_rev = s2["IS"]["revenue"]

Since the same snakeId system is used, any two companies can be compared using the same keys.

Financial Sector Specifics

Banks, insurance companies, securities firms, and other financial institutions have different financial statement structures from general manufacturers.

• revenue, cost_of_sales, current_assets, etc. may not exist
• A debt ratio exceeding 500% is normal
• Financial sector-specific accounts like interest_income, insurance_revenue exist

The insight engine automatically detects financial sector companies and applies separate criteria for analysis.