Previous

Page - 6

Calculation

Index can be constructed on both composite basis as well as sets of disaggregate indices for different cities and for different sectors in those cities. Having a set of disaggregate indices is much more relevant and useful for a diverse country like India and these set of disaggregate indices can always be used to form a composite nation wise index.

Above mentioned methods (hedonic method, repeat sales method and hybrid method) uses econometric regression methods to explain price levels or price changes and then uses the results to create an index of changes in price for a "typical" property. Thus, this method "holds constant" the quality of the property, a requirement for creating a price index for a heterogeneous good.

The fundamental relationship that is estimated is the link between the price of an asset and its characteristics. That is estimating the link between the transaction price of a property (commercial or residential) and characteristics such as its land area, structural area, quality of the structure, and location attributes.

When we apply the time series average to all the parameters identified above, the result is a set of valuations of each of the characteristics of properties in each time period. These time-varying valuations can then be applied to a particular set of property characteristics (often the sample's average values), yielding an estimate of property value for each time period. Index would be calculated monthly using a three-month moving average algorithm. Stock value weights would then be applied, along with the index divisor to arrive at each index level. For each reporting period, index points would be based on transactions collected in the data system on the first day of each of the preceding three months. Due to the nature of the real estate market, there is a natural lag between the transaction closing date and the data collection month.

On average, this lag is about three months (as seen in examples of US, China etc). For example, index number reported in September 2006 is based on transactions pulled down from the source system on June 1st, July 1st, and August 1st of 2006. Even though there would be lags in the reporting of data, this value would be labeled a September data point, with the majority of the closing dates of the transactions primarily in June, July and August. Hence, the data point reflects deals that closed primarily in these months of 2006. While there is a lag, the consistency of the methodology will accurately reflect the changes in real estate prices from month-to-month. The rolling moving average technique offsets lags in recorded sales from data providers, and maintains large sample sizes needed to produce statistically significant mean prices.

Next

* Contributed by: -
Rahul Abrol & Rahul Gulati,
MBA - IInd Year,
Institute of Management Technology, Ghaziabad.