search

berkleysayaka / DATA205-berkleysayaka

0 stars 0 forks source link

Business understanding ; identify project topics that I could research #3

Closed berkleysayaka closed 3 weeks ago

berkleysayaka commented 1 month ago
berkleysayaka commented 1 month ago

To analyze discrimination patterns in the DMV area, I want to focus on the following variables from HMDA data:

  1. Income: This can show whether or not there is discrimination by income when it comes to loaning money.
  2. Gender: This can assist in revealing the gender bias in lending.
  3. Race: This can be used to identify racial discrimination in the provision of mortgages.
  4. Place: Comparing the lending data for various areas in the DMV region will allow you to ascertain disparities geographically.
  5. Age: This can go a long way in explaining whether or not age-biased mortgage lending exists.
  6. Loan Agency (Company): Examining data by lender shows differences in the behaviors of different lender companies.
berkleysayaka commented 1 month ago

Updated variables names: Income: This can show whether or not there is discrimination by income when it comes to loaning money. Sex: This can assist in revealing the gender bias in lending. Race: This can be used to identify racial discrimination in the provision of mortgages. Places - Geographic unit: Comparing the lending data for various areas in the DMV region - Likely census tract/ county code Age: This can go a long way in explaining whether or not age-biased mortgage lending exists. Lender(Financial institution): Examining data by lender shows differences in the behaviors of different lender companies.

berkleysayaka commented 1 month ago

[HMDA Data: Identifying and Analyzing outliers ]

  1. HMDA data reveals disparities: Studies using HMDA data show that minorities are more likely to receive high-cost mortgages compared to non-Hispanic whites. This suggests potential bias in lending practices.
  2. HMDA data as a tool: While the data doesn't directly prove discrimination (it lacks details like credit scores), it can identify situations that warrant further investigation by the FDIC.
  3. Focus on fair lending: The FDIC uses HMDA data to enforce fair lending laws like the Equal Credit Opportunity Act (ECOA) and the Fair Housing Act (FHA).
  4. Loan review process: The FDIC conducts reviews based on initial HMDA data screening to identify potential lending disparities.

Overall, this webpage highlights the importance of HMDA data in promoting fair lending practices in the mortgage industry.

HMDA Data: Identifying and Analyzing Outliers | FDIC. www.fdic.gov/bank-examinations/hmda-data-identifying-and-analyzing-outliers#:~:text=Information%20collected%20under%20HMDA%2C%20including,to%20support%20implementation%20of%20the.

berkleysayaka commented 1 month ago

[Housing Finance Policy Center]

  1. Research: Conducts studies using HMDA data to examine issues related to fair lending, housing affordability, and mortgage market dynamics.
  2. Publications: Offers in-depth analyses and policy recommendations based on their research.
  3. Focus: Provides insights into the intersection of housing, finance, and public policy.

Housing Finance Policy Center | Urban Institute. 17 Oct. 2024, www.urban.org/policy-centers/housing-finance-policy-center.

berkleysayaka commented 1 month ago

[Detecting Racial Bias in Mortgage Lending Using Machine Learning]

  1. Predictive Modeling: Predicting loan default, interest rates, and other outcomes.
  2. Discrimination Detection: Identifying racial bias and other forms of discrimination in lending.
  3. Market Analysis: Segmenting the mortgage market, analyzing trends, and identifying opportunities.
  4. Regulatory Compliance: Assessing compliance with fair lending regulations and generating reports.
  5. Product Development: Designing new mortgage products and analyzing their performance.

“Fighting Discrimination in Mortgage Lending. |” MIT News | Massachusetts Institute of Technology, 30 Mar. 2022, news.mit.edu/2022/machine-learning-model-discrimination-lending-0330.

berkleysayaka commented 3 weeks ago

Choose three topics that I will be interested in research

  1. In-depth analysis of specific frauds. 1-1. Fraud patterns: In what cases are false income statements being made (e.g., self-employed, those with side jobs, etc.)? How fraudulent collateral valuations are carried out (e.g. fraudulent appraisals, overvaluing collateral properties, etc.) 1-2. Scale of damage: How much financing was fraudulently carried out through false income declarations. The amount of loss incurred by financial institutions due to fraudulent collateral valuations 1-3. Cause analysis: Why is this type of fraud so prevalent (e.g. lax screening standards, lenient penalties for fraud, etc.)

  2. Analysis of specific regions or attributes: Focus on specific regions or attributes such as low-income groups or minorities 2-1. Unequal impact: Are people from specific regions or attributes more likely to be victims of fraud 2-2. Are they being denied loans or forced to take out loans with high interest rates?

  3. Systemic failures: 3-1. Do existing regulations adequately protect people from certain regions or groups? 3-2. Socio-economic factors: How do poverty, education level, place of residence, etc. affect the incidence of fraud?