Is the proprietary Bank of America Bull & Bear Indicator, conventionally interpreted as good (bad) for future stock market returns when very low (very high), useful for market timing? Because the data series for this indicator is not publicly available, we ask Claude to estimate the data series from a publicly available chart. Claude generated estimates of beginning-of-month values of the indicator, with cautions as follows:
- This is a pixel reconstruction of a chart image, not sourced data, so data are approximate. Typical errors are ±0.2-0.3 units on a scale of 0-10 and ±2-4 weeks on turning points.
- June 2026 and July 2026 are set at 10.0, which look like contamination from chart annotations rather than genuine readings. [We use 9.4 and 9.7 from recent media reports rather than 10.0 for those two readings.]
- Anywhere annotations overlap the line, confidence is lower than elsewhere in the series.
Claude omitted June 2022 and July 2022 due to interfering annotations. Based on chart inspection, we used 0.0 for those two readings. With cautions in mind, we perform a battery of correlation, ranking and market timing tests on interactions between the estimated series and SPDR S&P 500 ETF Trust (SPY) returns. Using estimated monthly Bull & Bear Indicator values and end-of-month dividend-adjusted levels of SPY during February 2002 through July 2026, we find that:
Subscribe to Keep Reading
Get the research edge serious investors rely on.
- 1,200+ research articles
- Monthly strategy signals
- 20+ years of backtested analysis
Cancel anytime