You ask an aggregate question about the prepared 2022-2025 HMDA dataset.
HMDA Grouped 2022-2025
Prepared aggregate data
Read only
Turn your data into an analytics and LLM-ready environment.
Nationwide Mortgage Lending Analytics
Explore prepared Home Mortgage Disclosure Act (HMDA) data covering 2022-2025. Ask aggregate questions about mortgage applications, approvals, denials, loan purpose, applicant income, and lender activity.
Explore the annual source data in the FFIEC Data Browser.
Read-only aggregate dataset
How The Chatbot Works
Trusted metric rules and Terra planning translate the question into an allowed aggregate plan.
The server validates the plan, compiles read-only SQL, and queries the prepared MotherDuck table.
The chatbot returns a grounded answer, comparison chart, aggregate table, and executed SQL.
Example: "What is the approval rate by year?"
Example: "Show the denial rate by year."
Example: "Which loan purposes have the highest application volume?"
Data assistant
Ask the HMDA Analyst
Explore applications, approvals, funding, loan purpose, lender activity, and market trends.
GPT-5.6 Terra
Aggregate only
Suggested analysis
Continuous improvement
How real questions make the chatbot better
Questions the chatbot cannot answer well become signals for what to improve next. Each update is reviewed and tested before it reaches customers.
Real questions reveal the metrics, comparisons, and explanations people need.
A new, unclear, or unsupported question shows where the experience falls short.
Unanswered questions and feedback are grouped to find useful, repeated patterns.
Metric definitions, planning rules, prompts, and chart choices are refined.
Updates are checked for accuracy, safety, and clarity before they go live.
Clearer answers build trust and create the next round of useful questions.
Unanswered questionsShow missing capabilities and definitions.
Unclear answersReveal where explanations need work.
Repeated requestsHelp prioritize the most useful additions.