Chris Pollett >
Students > [Bio] [Blog] [MBS_BSF_linear regression Slides-PDF] |
CS297 ProposalHousing Market Crash Prediction Using Machine Learning and Historical DataParnika De (parnikade@gmail.com) Advisor: Dr. Chris Pollett Description: The housing market bubble burst caused the financial crisis of 2008. One of the main reasons for that was Collateral Debt Obligations (CDO). For the 2008 housing crisis, sub-prime mortgages played a huge role. Loans were given to people at high-interest rates who did not have collaterals (sub-prime loans). Then came the rating agencies who gave “AAA” rating to CDOs, even to the subprime CDO’s. We all know what a disaster it created. But that was 2008, a decade passed and banks learned from their mistakes and the US did not have a financial crisis since the US economy is stronger than ever. But is it going to be all good forever? The answer is we don’t know. Therefore in this project, I aim to analyze US housing market data and other financial data to predict whether everything is good or are we heading south. To do this project, I would use a forecasting model to analyze the data and build an ML model that would help me predict “The future of the housing market”. Schedule:
Deliverables: The full project will be done when CS298 is completed. The following will be done by the end of CS297: 1. Data preparation(Cleaning of Data set) 2. Code HMM on a small dataset 3. Learn about Long short term Memory and code it 4. Apply Linear Regression on the Housing dataset 5. CS 297 report References: [2017] The Great Recession: A Macroeconomic Earthquake. Lawrence J. Christiano. Federal Reserve Bank of Minneapolis. 2017. [2011] The Role of ABS, CDS and CDOs in the Credit Crisis and the Economy. Robert A. Jarrow. 2011 [2016] The 2007–2009 Financial Crisis: An Erosion of Ethics: A Case Study. Edward J. Schoen. Journal of Business Ethics. 2016 [2009] Financial crises and bank failures: A review of prediction methods. Yuliya Demyanyk, Iftekhar Hasan . Elsevier Omega. 2009. |