Finance
21 columns
200 rows
8 formats

AI-Driven Loan Default Scenarios

This synthetic dataset provides detailed, time-stamped borrower and loan records, enriched with macroeconomic scenario variables for robust loan default prediction modeling. It is designed for fintech machine learning experiments, enabling analysis of borrower behavior and loan performance under varied economic conditions. The dataset supports scenario-based risk modeling, stress testing, and algorithm benchmarking.

last updated 12 Jul 2025
by GoMask

The brief that made it

Training and benchmarking loan default prediction models

Sample rows

preview · 8 of 200 rows · all 21 columns

borrower_idstring	economic_scenariostring	interest_ratefloat	defaultedboolean	loan_purposestring	loan_idstring	snapshot_datedate	ageinteger	employment_statusstring	annual_incomefloat	credit_scoreinteger	loan_amountfloat	loan_term_monthsinteger	current_balancefloat	payment_due_datedate	days_past_dueinteger	num_past_due_paymentsinteger	default_datedate	unemployment_ratefloat	inflation_ratefloat	regionstring
BRW00001	expansion	6.2	false	auto	LN00001	2023-05-10	29	employed	78500	712	25000	60	19500	2023-06-09	0	0	blank	3.9	2.4	USA
BRW00002	recession	39.8	true	medical	LN00002	2022-12-15	62	retired	27000	388	8000	24	6730	2023-01-14	38	2	2023-01-24	16.2	9.1	Greece
BRW00003	recovery	8.4	false	education	LN00003	2023-02-01	20	student	9000	540	6500	36	5400	2023-03-01	0	0	blank	11.4	2.7	India
BRW00004	baseline	4.3	false	home_improvement	LN00004	2023-07-22	42	employed	112000	771	47000	84	39250	2023-08-21	0	0	blank	4.8	2.2	Canada
BRW00005	expansion	7.1	false	business	LN00005	2022-11-30	36	self-employed	56500	679	31000	60	29500	2022-12-30	0	0	blank	6.1	1.7	Singapore
BRW00006	crisis	78.5	true	medical	LN00006	2023-04-14	58	retired	18000	334	3500	12	3270	2023-05-14	75	3	2023-05-30	21.7	13.2	Venezuela
BRW00007	baseline	5.5	false	auto	LN00007	2023-03-17	34	employed	102700	742	40000	60	38700	2023-04-16	0	0	blank	3.2	1.1	Germany
BRW00008	expansion	9.9	false	business	LN00008	2022-10-05	47	self-employed	42000	613	12000	36	10400	2022-11-04	0	0	blank	5.9	2	UK
BRW00009	crisis	92.6	true	medical	LN00009	2023-06-24	56	retired	11800	382	2500	6	2500	2023-07-23	56	2	2023-08-02	31.2	22.4	Zimbabwe
BRW00010	recovery	10.5	false	education	LN00010	2023-01-12	18	student	4200	362	2000	12	1200	2023-02-11	0	0	blank	13.1	2.6	Brazil
BRW00011	crisis	92	true	debt_consolidation	LN00011	2023-02-25	38	unemployed	0	410	1400	6	1400	2023-03-27	74	2	2023-04-01	22.7	14.3	Nigeria
BRW00012	expansion	8.1	false	home_improvement	LN00012	2023-05-05	31	employed	52000	680	16500	36	12100	2023-06-04	0	0	blank	4.2	1.5	Australia
BRW00013	baseline	7.9	false	education	LN00013	2023-04-19	23	student	11950	629	10000	60	9800	2023-05-19	0	0	blank	8.7	2.3	South Africa
BRW00014	expansion	3.7	false	home_improvement	LN00014	2023-03-03	44	employed	127500	808	58000	84	57950	2023-04-02	0	0	blank	2.8	2.1	Switzerland
BRW00015	recovery	21.3	false	business	LN00015	2023-01-27	53	self-employed	29750	497	11000	24	5270	2023-02-26	0	0	blank	13.9	5.1	Turkey
BRW00016	baseline	9.2	false	auto	LN00016	2022-12-19	27	employed	47100	667	13000	48	11750	2023-01-18	0	0	blank	4.4	3	UAE
BRW00017	expansion	3.9	false	home_improvement	LN00017	2023-05-22	39	employed	108300	795	51000	60	47900	2023-06-21	0	0	blank	3.9	2.6	USA
BRW00018	baseline	8.7	false	auto	LN00018	2023-02-08	41	employed	48000	654	17500	48	14980	2023-03-10	0	0	blank	5.7	2.7	France
BRW00019	crisis	98.6	true	medical	LN00019	2023-07-09	77	retired	15400	320	1800	6	1800	2023-08-08	35	2	2023-08-18	42	27.5	Syria
BRW00020	baseline	13.6	false	business	LN00020	2023-05-18	48	self-employed	33000	582	6500	24	2500	2023-06-17	0	0	blank	8.3	2.4	Italy

What the 200 rows show

from the 200-row sample

Crisis (economic scenario) stands out: mean interest_rate is 86.5, against 12.9 for the rest.

26%defaulted = true
8.4median interest_rate
5employment statuses
48regions
38median age
41,500median annual_income

Mean interest_rate by economic_scenario200 rows

interest_rate200 rows, in bands of 10

Every row with interest_rate of 30 or more has defaulted = true. None below it.

loan_purpose200 rows · 7 values

home_improvement51
medical36
auto34
education31
business24
debt_consolidation17
other7

21 columns by typefrom the column list below

string 6
integer 5
float 6
date 3
boolean 1

Columns

21 columns in four groups

blueprint · 21 columns
column	type	description	example
Text 6 columns
`borrower_id`	string	Unique identifier for each synthetic borrower.unique	BRW00001
`loan_id`	string	Unique identifier for each loan issued to a borrower.unique	LN00001
`employment_status`	string	Current employment status of the borrower.employed · unemployed · self-employed · retired · student	employed
`loan_purpose`	string	Purpose for which the loan was taken.7 values	auto
`economic_scenario`	string	Simulated macroeconomic scenario affecting the borrower's environment.baseline · recession · expansion · crisis · recovery	expansion
`region`	string	Geographical region or country of the borrower.	USA
Numbers 11 columns
`age`	integer	Age of the borrower at the time of loan origination.18 to 100	29
`annual_income`	float	Annual income of the borrower in USD.0 or more	78500
`credit_score`	integer	Borrower's credit score at the time of loan application.300 to 850	712
`loan_amount`	float	Principal amount of the loan in USD.1,000 or more	25000
`loan_term_months`	integer	Duration of the loan in months.6 to 360	60
`interest_rate`	float	Annual interest rate applied to the loan (as a percentage).0 to 100	6.2
`current_balance`	float	Outstanding loan balance at the snapshot date.0 or more	19500
`days_past_due`	integer	Number of days the payment is overdue as of the snapshot date.0 or more	0
`num_past_due_payments`	integer	Total number of past due payments for this loan.0 or more	0
`unemployment_rate`	float	Unemployment rate (%) in the borrower's region at snapshot date.0 to 100	3.9
`inflation_rate`	float	Inflation rate (%) in the borrower's region at snapshot date.-10 to 100	2.4
Dates and times 3 columns
`snapshot_date`	date	Date of the recorded features and scenario snapshot.	2023-05-10
`payment_due_date`	date	Next payment due date for the loan.	2023-06-09
`default_date`	date	Date when the loan was marked as defaulted (null if not defaulted).optional	2023-01-24
True or false 1 column
`defaulted`	boolean	Whether the borrower has defaulted on the loan as of the snapshot date.	false

Use it for

A finance dashboard
The defaulted rate, interest_rate by economic_scenario and a breakdown of loan_purpose. Excel, Power BI or Tableau.
Why do the 39 crisis rows have a mean interest_rate of 86.5?
A root-cause class exercise
Hand out the rows and one question. The answer is in the data, not in the brief.
Borrowers200
BRW000016.2expansion
BRW0000239.8recession
BRW0000678.5crisis
A software demo
Believable borrowers with loan_id, snapshot_date and age to fill a screen in front of a buyer.

This dataset200 rows21 columns

Yours1,000 rows21 columnsregion: UK only

blueprint · ai-driven-loan-default-scenarios

Behind this dataset

Same schema. As many rows as you need.

These 200 rows came out of a blueprint — 21 columns with generation rules behind each one. Open it in Data Factory to retune a column, add your own, wire in foreign keys, and run it at the size you actually need.

Rules it was built with

Borrower profiles anonymized
Payment events simulated monthly
Defaults influenced by synthetic unemployment rates
Missing payment probability modeled per profile

Rows1,000 rows10,000 rows100,000 rows1,000,000 rows

Open the blueprint in Data Factory

1 credit per row. Preview 20 rows free. Buy rows for the full run.

Exports: CSV, JSON, JSONL, Parquet, SQL, Excel, TSV, XML
Licence: yours to use, including commercially
API slug: ai-driven-loan-default-scenarios

What should your data show?

No signup. No card.

Sample rows

What the 200 rows show

Columns

Use it for

A finance dashboard

A root-cause class exercise

A software demo

Make it yours.

Same schema. As many rows as you need.

More in Finance

What should your data show?