An R package for calculating repeat-sale price indices

Abstract

Residential property price indices are valuable indicators of economic activity and financial stability. Calculating pure, quality adjusted price changes using real estate data presents a number of obstacles, however, as properties each have a unique set of price determining characteristics. The repeat-sales method offers a means to construct a constant-quality price index by exploiting multiple sales for the same property over time, although calculating reproducible results on an ongoing basis following this methodology is not trivial. This paper outlines an R package developed at Statistics Canada for calculating a variety of repeat-sale price indices found in the literature and used in practice, as well as supporting tools to get price data into a form that is suitable for calculating an index. The package is designed to be simple, easy to use, and suitable for operational use, without requiring much prior knowledge of the repeat-sales approach or R.

Keywords

• price index
• repeat-sales
• case-shiller
• r