Dechow F-Score

Calculate the Dechow F-Score, the probability that a company’s financial statements are misstated relative to the average company. It was estimated on companies the SEC charged with misstating their earnings, and combines:

• Accrual quality: RSST accruals (the change in non-cash working capital, non-current operating assets and financial assets), the changes in receivables and inventory, and the share of soft assets (assets other than cash and short-term investments and net property, plant and equipment).
• Performance: the change in cash sales and in return on assets.
• Incentives: whether the company issued long-term debt or stock.

Predicted Value = -7.893 + 0.790 RSST Accruals + 2.518 ΔReceivables + 1.191 ΔInventory + 1.979 Soft Assets + 0.171 ΔCash Sales - 0.932 ΔROA + 1.029 Issuance Probability = e^PV / (1 + e^PV), F-Score = Probability / 0.0037

An F-Score of 1 is average risk; above 1 the risk is elevated, above 1.85 substantial and above 2.45 high.

For more information, see the following paper:

• Dechow, P.M., Ge, W., Larson, C.R. and Sloan, R.G. (2011). “Predicting Material Accounting Misstatements.” Contemporary Accounting Research, 28(1), 17-82.

Also known as: F-Score (Dechow), misstatement score, earnings manipulation probability.

Dechow F-Score in Python or via the MCP

Install the Finance Toolkit with:

pip install financetoolkit -U

Then call get_dechow_f_score as shown below.

from financetoolkit import Toolkit

toolkit = Toolkit(
    ["AAPL", "MSFT", "SMCI"],
    api_key="FINANCIAL_MODELING_PREP_KEY",
    start_date="2021-01-01",
    end_date="2025-12-31",
)

toolkit.models.get_dechow_f_score().xs("F-Score", level=1)

Or, after setting up the Finance Toolkit MCP, ask:

Which returns:

Parameters

get_dechow_f_score accepts the following parameters.

Related Models