Leveraging unique behavioral data to create powerful market indices

In many fields of economics and finance it is useful to be able to compare the trends and performance of particular market sectors to one another.

In order to do this, a series of indices can be created to measure in relative terms the change in the value of, or size of, a particular segment of the market. These indices can be tracked over time from the initial base period, and peaks, troughs, and changes in growth can be analyzed to better understand the impact of economic conditions on various markets.

When it comes to financial markets, an index typically measures the performance of a basket of securities that represent a particular market sector. For example an index may capture securities in a particular industry sector e.g. Technology, or even the entire market such as the Dow Jones Industrial Average, or the S&P 500. These indices measure the price performance of the securities contained within them in a standardized way.

Indices are also used in many other areas such as inflation, property prices, interest rates, business activity, consumer or business confidence. These all provide a means by which to understand the performance of a part of a market or economy in a consistent and reliable manner.

Indices can also be a useful means by which to extract signals from the underlying data (e.g. share prices) in situations where it is not possible to share the underlying pricing data. There may be data licensing or privacy limitations in sharing the more granular data, where an index avoids disclosure of the underlying data however still provides a valuable insight into the market in question.

These all provide a means by which to understand the performance of a part of a market or economy in a consistent and reliable manner.

It is typical to generate a headline index at a macro level (i.e. whole market or geographic region) and to then supply indices at lower market segments or geographic regions for more detailed analysis. Indices at a lower level of geography, say, can be used as a means to benchmark sales performance of your company against the whole market to identify areas where growth in sales is below that of the overall market.

An example of this is the recently released PEXA Refinance index, which RoZetta Technology was proud to develop in partnership with PEXA.

The index measures the performance of the mortgage refinance market both in terms of the volume of activity in the market, and the median value of this refinance activity. Each state can be compared to the national trend and each other to identify which states are outperforming others. Similarly, this can then be done at lower geographic levels, which can form the basis of a benchmark with which a lender, say, might compare the growth they are seeing in their refinance activity compared to the market. Further analysis of the local area may reveal a clearer picture of any emerging underlying risk such as a concentration of activity in a particular area or could help assist how to best allocate resources (branches, lending staff) to a given area to best service the needs of their customers. Importantly, if changes are made as a result of this analysis, the index serves again as a useful benchmark to measure the impact to understand if the action has been successful.

To generate an index does require specialist expertise to ensure that:

• Movement in the Index is caused by the economic or market activity that is being tracked rather than the composition of underlying data being included in the index (e.g. share prices, trades, refinance transactions)
• Outliers are handled appropriately so as not to perturb the index
• Changes in the composition of the underlying elements of the index need to be managed appropriately
• The index method used needs to ensure that turning points in trends are handled appropriately (i.e. don’t lag, the index isn’t overly sensitive to turning points)