Abstract
House price indices are difficult to compute because houses are essentially unique, nonreplicable goods and transactions are relatively infrequent. Furthermore, unlike in organized markets, there is considerable opacity concerning transaction prices.The traditional real estate agent coexists nowadays with Web sites, from which information can be obtained cheaply and easily. This paves the way for new approaches which tap this resource to produce timely, quality and location adjusted housing price indices. We describe one such approach and show R to be a tool of choice in every step of the implementation.
Original language | English |
---|---|
Title of host publication | Data Mining Applications with R |
Publisher | Academic Press |
Pages | 273-297 |
Number of pages | 25 |
ISBN (Print) | 9780124115118 |
DOIs | |
Publication status | Published - 1 Dec 2013 |
Externally published | Yes |
Keywords
- Geographically weighted regression
- Housing prices
- Price indices
- Semiparametric models