Skip to content
All library documents

Finding ETF Benchmarks and Building Longer Return Histories

Article Quant Q&A · Author: Elrond
Read the original

Summary

This discussion addresses how to identify the declared benchmark for US-listed ETFs and retrieve benchmark histories for backtesting. One Bloomberg workflow is to screen funds with the ETF function, export the results, then use the primary benchmark field to obtain the index ticker and download its time series through the Excel interface. The benchmark can serve as a proxy for the ETF over periods where the fund’s own history is too short, though the document does not specify when that substitution is appropriate.

The responses note practical data issues that can distort comparisons. ETF prices may be stale on less liquid exchange listings, so exchange codes matter; net asset value can be an alternative comparison series. Asian and emerging-market funds also involve currency conversion, which can create tracking differences. Another reply points to index constituents as a way to filter data, but does not provide a comprehensive method for generating a complete ETF-to-benchmark mapping. The questioner reports that vendor fields were often missing or incorrect, so benchmark identifiers should be checked before using them in research.

Key ideas

  • Bloomberg’s ETF screening and primary benchmark field can be used to map funds to benchmark tickers.
  • Benchmark index histories may help extend a fund’s backtest, subject to suitability checks.
  • Exchange listing selection matters because illiquid ETF listings can contain stale prices.
  • NAV data may be useful when market-price histories are sparse or misleading.
  • Currency conversion can contribute to tracking differences for Asian and emerging-market ETFs.

Tags

EquitiesBacktesting
Full text
# Database of ETF and underlying index from Thomson Reuters or Bloomberg


# Database of ETF and underlying index from Thomson Reuters or Bloomberg












I want to generate a table of ETFs listed in the US and declared benchmarks. The idea is that I want to have the possibility to use the benchmark as a proxy of the ETF to run longer backtests when appropriate.

Among cheap data vendors, there is a nice table provided by ETF Global on Quandl that does the job. However, there is no ISIN or SEDOL for the ETFs and no identifiers for the benchmark, making not possible to automatically download the corresponding total return time series from Thomson Reuters or Bloomberg afterwards.

I've spent hours with Eikon, Datastream and chatting with the support to create this table using Thomson Reuters data, but many benchmarks are wrong or missing. I do not have access to Bloomberg.

Question: is it possible to retrieve such table from Bloomberg or Thomson Reuters?

## Answer by HLus (score 3)

https://quant.stackexchange.com/a/35528

You can find the underlying data for the big indices, such as S&P 500, on our website www.simfin.com, freely available and instantly downloadable as excel.

To this date, we have financial ratios, Financial statements (directly sourced from the SEC's XBRL data and up to 10y back) and stock prices for over 1000+ US companies, which hopefully might provide you some valuable info on generating your table.

For your case, it will be advisable to filter the data regarding the index membership.

## Answer by vanguard2k (score 1)

https://quant.stackexchange.com/a/35535

So here's the answer for Bloomberg I suppose plus some extra advice when looking at the data you get from there.

There is the function ETF that lets you screen for ETFs. Unfortunately, I have not found a possibility to screen by benchmark.

Take AGG US Equity as an example. There is the field "FUND_BENCHMARK_PRIM" which points to the BM ticker. (This is what you see when you use the COMP function). With the help of this ticker, you can easily download the bm timeseries.

So what you could try:

- Export the filter results from the ETF function into excel

- Download the BM ticker via the Excel BB API

A couple of warnings: If you compare the BM returns to the one of the ETF on Bloomberg via ticker, be aware that the Exchange code DOES matter. On some of them, the ETF is scarcely traded and as a result there are lots of stale prices in Bloomberg. Of course, you could still sell this on the secondary market. Just choose a different exchange with more liquid prices to compare. Or you could look at the NAV instead (also, if you NAV trade the ETFs of course). If you look at Asian or EM ETFs, there is the currency conversion that gives you lots of TE so its better to look twice before jumping to conclusions.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.