metadata
pretty_name: Stocks Weekly CFPB Complaints
language:
- en
license: other
task_categories:
- time-series-forecasting
- tabular-regression
tags:
- finance
- quantitative-trading
- backtesting
- algorithmic-trading
- stocks
- equities
- weekly
size_categories:
- 1M<n<10M
extra_gated_prompt: >-
This dataset is free to browse and gated for download. Approval is tied to a
Papers With Backtest subscription, which also covers the other datasets in
this organisation and the strategy catalogue at
https://paperswithbacktest.com. Plans and what each one includes:
https://paperswithbacktest.com/pricing
dataset_info:
features:
- name: symbol
dtype: string
- name: datetime
dtype: string
- name: company
dtype: string
- name: bloomberg_share_id
dtype: string
- name: culpability_score
dtype: float64
- name: complaint_score
dtype: float64
- name: grievance_score
dtype: float64
- name: total_risk_rating
dtype: float64
- name: product
dtype: string
- name: sub_product
dtype: string
- name: issue
dtype: string
- name: sub_issue
dtype: string
- name: consumer_complaint_narrative
dtype: string
- name: company_public_response
dtype: string
- name: state
dtype: string
- name: zip_code
dtype: string
- name: tags
dtype: string
- name: consumer_consent_provided
dtype: string
- name: submitted_via
dtype: string
- name: date_sent_to_company
dtype: string
- name: company_response_to_consumer
dtype: string
- name: timely_response
dtype: string
- name: consumer_disputed
dtype: string
- name: selected_name
dtype: string
- name: similarity
dtype: float64
splits:
- name: train
num_examples: 1730498
Stocks Weekly CFPB Complaints
Weekly mapping of CFPB consumer complaints to listed companies with risk scores and metadata.
1,730,498 rows over 1,283 symbols, 25 columns, covering 2011-12-01 to 2026-04-10. The most recent observation is 2026-04-10.
Why It Matters
This dataset introduces consumer-complaint signals into financial equity models by:
- Reputational risk: Rising complaint intensity can foreshadow regulatory actions and spread widening for financial issuers.
- Event detection: Weekly cadence captures fresh issues that may not yet be reflected in prices.
- Alternative sentiment: Complaint narratives add a consumer-sentiment layer beyond traditional news feeds.
Load It
Installation/Upgrade:
pip install --upgrade pwb-toolbox
Load the Dataset:
from pwb_toolbox import datasets as pwb_ds
df = pwb_ds.load_dataset("Stocks-Weekly-CFPBComplaints", symbols=["EFX"])
print(df.iloc[0, :])
Example Output:
symbol EFX
datetime 2025-11-15 00:00:00
company equifax, inc.
bloomberg_share_id BBG001SXXC92
culpability_score 0.157216
complaint_score 0.02616
grievance_score 0.054874
total_risk_rating 0.079417
product Credit reporting or other personal consumer re...
sub_product Credit reporting
issue Incorrect information on your report
sub_issue Information belongs to someone else
consumer_complaint_narrative None
company_public_response None
state CA
zip_code 94579
tags None
consumer_consent_provided None
submitted_via Web
date_sent_to_company 2025-11-15
company_response_to_consumer In progress
timely_response Yes
consumer_disputed None
selected_name equifax, inc.
similarity 1.0
Columns
| Column Name | Description |
|---|---|
| symbol | Stock ticker linked to the complaint. |
| datetime | Week-ending snapshot date (YYYY-MM-DD). |
| company | Reported company name in the complaint. |
| bloomberg_share_id | Bloomberg share identifier for the matched company. |
| culpability_score | Model score estimating company culpability. |
| complaint_score | Model score reflecting complaint severity. |
| grievance_score | Model score capturing grievance intensity. |
| total_risk_rating | Aggregated risk rating from complaint signals. |
| product | Product category mentioned in the complaint. |
| sub_product | Sub-product category for the complaint. |
| issue | Primary issue raised. |
| sub_issue | Detailed sub-issue for the complaint. |
| consumer_complaint_narrative | Narrative text describing the complaint. |
| company_public_response | Company's public response text, if available. |
| state | U.S. state of the complainant. |
| zip_code | ZIP code of the complainant. |
| tags | Tags applied to the complaint. |
| consumer_consent_provided | Consent flag for sharing the complaint narrative. |
| submitted_via | Channel through which the complaint was submitted. |
| date_sent_to_company | Date the complaint was sent to the company. |
| company_response_to_consumer | Company's response classification. |
| timely_response | Indicates whether the response was timely. |
| consumer_disputed | Whether the consumer disputed the resolution. |
| selected_name | Matched entity name used to link the complaint to the symbol. |
| similarity | Matching similarity score. |
Access
Browsing the card and the schema is open to anyone. Downloading the files needs an approved request, tied to a subscription: what each plan includes. The same subscription covers the other datasets in this organisation.
Elsewhere
- Dataset page and coverage charts
- The strategy catalogue, 3,806 papers and 4,837 replicated strategies
pwb-toolbox, the loader used in the snippet aboveawesome-systematic-trading, the replicated strategies with their measured Sharpe- Every dataset in this organisation
Papers With Backtest publishes 32 datasets on the Hub and codes the papers that use them. Every strategy in the catalogue is run over its own full history before it is published, which is where the numbers above come from.