Why Price-to-Sales Alone Is a Weak Value Signal

Summary

The document considers whether the price-to-sales ratio can predict broad stock-market movements or serve as a measure of intrinsic value. A questioner reports substantial explained variance from a regression using P/S and suggests that market gains may account for returns tracking the market. The response disputes the inference that the ratio is a reliable predictor, both for an index and for individual stocks, and questions whether the reported regression was set up correctly.

The answer also cautions that P/S is not an unambiguous measure of intrinsic value. Value strategies commonly combine several attributes, such as price-to-book and price-to-cash-flow, because any single measure has limited predictive power. No regression specification, dataset, or independent empirical results are provided, so the exchange offers a methodological caution rather than a tested estimate of P/S performance. Researchers should scrutinize the target, timing, and construction of any reported relationship before treating it as a forecasting signal.

Key ideas

• A reported regression fit does not by itself establish that P/S predicts future returns.
• The response questions the setup behind the stated explained variance.
• Price-to-sales is a subjective and incomplete proxy for intrinsic value.
• Value approaches often combine multiple valuation attributes because individual measures have limited predictive power.

Tags

# Is it pretty easy to predict the stock market from price/sales?


# Is it pretty easy to predict the stock market from price/sales?












Except the recent boom (entirely driven by tech) isnt the stock market just an oscillator around quarterly price to sales? I'm getting like 25% explained variance just from that.

The returns are close to market but that's just because most quarters have been up.

## Answer by Chris (score 1)

https://quant.stackexchange.com/a/57212

Meaning, 'as a measure of intrinsic value, cant you use P/S to predict, loosely, future stock price movements'?

Not even close at the index level, and certainly not at the stock level. Not sure how you've set up your regression to get an R2 of .25 with P/S as your only independent variable, but I'd almost guarantee there's an issue with your calculation.

Even asserting P/S is a robust estimator of intrinsic value is subjective. Most factor definitions of value use a combination of attributes (inclusive of P/S, P/B, P/CF, among others) because individual attributes have limited predictive power.

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.