System and method for providing a regularized adjusted weighted repeat sale index
View Patent ↗Systems, methods, and computer-readable storage media are described for estimating real estate property values based on an adjusted repeat sales model using a regularized estimator. In one exemplary embodiment, a computer-implemented method calculates data for estimating the adjustments from an aggregated level to a first disaggregated level by marking a first transaction to a second transaction using a repeat sales house price index function at the aggregated level. The method also determines, using the calculated data, a regularized estimate of the deviation between the repeat sales house price index at the aggregated level and a repeat sales house price index at the first disaggregated level. The method further calculates the repeat sales house price index at the first disaggregated level based on the determined regularized estimate of the deviation from the aggregated level.
1. A computer-implemented method for estimating a weighted repeat sales index using a regularized estimator using a computer system that includes a processor, and a database, the method comprising the following steps:
calculating, with the processor using information stored in the database, data for estimating adjustments from an aggregated level to a first disaggregated level by marking a first transaction to a second transaction using a repeat sales house price index function at the aggregated level;
combining, using the processor, one or more repeat transaction returns into a first set, the one or more repeat transaction returns being based on a price of the first transaction price and a price of the second transaction;
determining, using the processor, a second set of one or more regressors based on at least one of a time period the first transaction took place, a time period the second transaction took place, the first transaction being a refinance transaction or not a refinance transaction, or the second transaction being a refinance transaction or not a refinance transaction;
determining, with the processor, a regularized estimator in a linear fashion based on the first set and the second set, and storing the regularized estimator in the database;
determining, with the processor, using the calculated data and the regularized estimator, a regularized estimate of the deviation between the repeat sales house price index at the aggregated level and a repeat sales house price index at the first disaggregated level; and
calculating, with the processor, the repeat sales house price index at the first disaggregated level based on the determined regularized estimate of the deviation from the aggregated level.
2. The computer-implemented method of claim 1 , further comprising the step of:
calculating a repeat sales house price index at a second disaggregated level based on a determined regularized estimate of the deviation from the first disaggregated level.
3. The computer-implemented method of claim 2 , further comprising the step of:
recalculating a repeat sales house price index at the first disaggregated level based on a weighted average of a determined regularized estimate of the deviation from the second disaggregated level.
4. The computer-implemented method of claim 3 , further comprising the step of:
recalculating a repeat sales house price index at the aggregated level based on a weighted average of a determined regularized estimate of the deviation from the first disaggregated level.
5. The computer-implemented method of claim 1 , wherein the aggregated level is a geographic region within the United States of America or a state.
6. The computer-implemented method of claim 1 , wherein the first disaggregated level is a county, and the second disaggregated level is a zip code area.
7. The computer-implemented method of claim 1 , wherein the repeat sales house price index function at the aggregated level, the first disaggregated level, and the second disaggregated level are weighted to differentiate between purchase transactions and refinance transactions.
8. A system for estimating a weighted repeat sales index using a regularized estimator, the system comprising:
a processor;
a database; and
a memory containing instructions which, when executed, causes the system to perform a method comprising:
calculating, with the processor using information stored in the database, data for estimating adjustments from an aggregated level to a first disaggregated level by marking a first transaction to a second transaction using a repeat sales house price index function at the aggregated level;
combining, using the processor, one or more repeat transaction returns into a first set, the one or more repeat transaction returns being based on a price of the first transaction price and a price of the second transaction;
determining, using the processor, a second set of one or more regressors based on at least one of a time period the first transaction took place, a time period the second transaction took place, the first transaction being a refinance transaction or not a refinance transaction, or the second transaction being a refinance transaction or not a refinance transaction;
determining, with the processor, a regularized estimator in a linear fashion based on the first set and the second set, and storing the regularized estimator in the database;
determining, with the processor, using the calculated data and the regularized estimator, a regularized estimate of the deviation between the repeat sales house price index at the aggregated level and a repeat sales house price index at the first disaggregated level; and
calculating, with the processor, the repeat sales house price index at the first disaggregated level based on the determined regularized estimate of the deviation from the aggregated level.
9. The system of claim 8 , further comprising:
calculating a repeat sales house price index at a second disaggregated level based on a determined regularized estimate of the deviation from the first disaggregated level.
10. The system of claim 9 , further comprising:
recalculating a repeat sales house price index at the first disaggregated level based on a weighted average of a determined regularized estimate of the deviation from the second disaggregated level.
11. The system of claim 10 , further comprising:
recalculating a repeat sales house price index at the aggregated level based on a weighted average of a determined regularized estimate of the deviation from the first disaggregated level.
12. The system of claim 8 , wherein the aggregated level is a geographic region within the United States of America or a state.
13. The system of claim 8 , wherein the first disaggregated level is a county, and the second disaggregated level is a zip code area.
14. The system of claim 8 , wherein the repeat sales house price index function at the aggregated level, the first disaggregated level, and the second disaggregated level are weighted to differentiate between purchase transactions and refinance transactions.
15. A computer-readable medium including program instructions for performing, when executed by computer system that includes a processor, a real estate database, a display, and a memory, a method for estimating a weighted repeat sales index using a regularized estimator, the method comprising:
calculating, with the processor using information stored in the database, data for estimating adjustments from an aggregated level to a first disaggregated level by marking a first transaction to a second transaction using a repeat sales house price index function at the aggregated level;
combining, using the processor, one or more repeat transaction returns into a first set, the one or more repeat transaction returns being based on a price of the first transaction price and a price of the second transaction;
determining, using the processor, a second set of one or more regressors based on at least one of a time period the first transaction took place, a time period the second transaction took place, the first transaction being a refinance transaction or not a refinance transaction, or the second transaction being a refinance transaction or not a refinance transaction;
determining, with the processor, a regularized estimator in a linear fashion based on the first set and the second set, and storing the regularized estimator in the database;
determining, with the processor, using the calculated data and the regularized estimator, a regularized estimate of the deviation between the repeat sales house price index at the aggregated level and a repeat sales house price index at the first disaggregated level; and
calculating, with the processor, the repeat sales house price index at the first disaggregated level based on the determined regularized estimate of the deviation from the aggregated level.
16. The computer-readable medium of claim 15 , further comprising:
calculating a repeat sales house price index at a second disaggregated level based on a determined regularized estimate of the deviation from the first disaggregated level.
17. The computer-readable medium of claim 16 , further comprising:
recalculating a repeat sales house price index at the first disaggregated level based on a weighted average of a determined regularized estimate of the deviation from the second disaggregated level.
18. The computer-readable medium of claim 17 , further comprising:
recalculating a repeat sales house price index at the aggregated level based on a weighted average of a determined regularized estimate of the deviation from the first disaggregated level.
19. The computer-readable medium of claim 15 , wherein the aggregated level is a geographic region within the United States of America or a state.
20. The computer-readable medium of claim 15 , wherein the first disaggregated level is a county, and the second disaggregated level is a zip code area.
21. The computer-readable medium of claim 15 , wherein the repeat sales house price index function at the aggregated level, the first disaggregated level, and the second disaggregated level are weighted to differentiate between purchase transactions and refinance transactions.