Checking the Provenance of Early Daily S&P 500 Index History

Summary

The document questions the reliability and construction of historical daily S&P 500 index data before the modern index existed. It notes that one source claims a much longer history, including daily observations from 1927, and wonders whether those early daily values represent an earlier, broader index or were derived by interpolating monthly data. It also observes a small first-lag autocorrelation in exploratory analysis and asks whether interpolation could explain it.

These are data-provenance and time-series diagnostics relevant to historical index modeling. The document does not identify the source, verify its methodology, compare it with independent records, or determine whether the autocorrelation is artificial or reflects historical market behavior. Its dates and autocorrelation figure are reported as part of the question, not as validated findings. Any backtest using such early observations would need independent confirmation of the index construction and sampling process.

Key ideas

• The document questions whether claimed early daily index history was observed or reconstructed from lower-frequency data.
• It distinguishes the earlier index composition from the later 500-stock index.
• A small first-lag autocorrelation is noted as a possible interpolation artifact or historical market feature.
• The source and data-generation method are not verified in the document.

Tags

# Is there daily SPX level data going back to 1927?


# Is there daily SPX level data going back to 1927?












While attempting to model the SPX index over time, I found a source here that purportedly has historical daily SPX data going back to 1789 which very likely seems to be backcasted since the ~500 stock version only came into being around 1957.

What I am more uncertain about is if the daily data from this source dating back to 1927 (when SPX was a ~200(?) stock index) is simply a linearly interpolated version of the monthly data at this time? since I have not seen many data sources claim to have daily data for SPX starting in 1927.

Additionally, when doing some exploratory data analysis, I found that there was notable, but still small autocorrelation (~0.1) at the first lag which I know could be from the market being less efficient during this period, but still seemed like it could be an artifact of interpolation.

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.