Map ordinary 12-month outcomes before stacking valuation, rates, and seasonality

Place a single index position in a broader regime

Using valuation, interest-rate, and calendar relationships together places a single index position inside a broader market regime rather than judging it in isolation.

Editorial reading: treat the position as a regime decision. First map ordinary rolling 12-month outcomes. Then stack starting earnings yield, the rate backdrop, and calendar context before asking whether the trade sits in a historically richer or thinner return bucket.

First map ordinary 12-month outcomes

The rolling 12-month change is the overlapping year-over-year percentage move in the index. It defines ordinary outcomes and is also the later result being sorted.

Across overlapping 12-month windows from 1945 through May 2007, the index's average price change excluding dividends was 8.4 percent, with a largest gain of 53.4 percent and a largest decline of 41.4 percent.

In that same sample the median 12-month change was 9.5 percent, three-quarters of windows were better than a 1.9 percent decline, and only one-tenth exceeded a 28.6 percent gain.

A 15.7 percent standard deviation of those 12-month changes understates tail risk: a 41.4 percent decline would be extremely rare under a normal curve, yet it appeared inside the 62-year sample.

Then stack starting earnings yield

Earnings yield is trailing reported earnings divided by the index price, the inverse of the price/earnings ratio. It is used here as a starting valuation state that is known at the time, so the later return is not mixed into the starting signal.

Correlation analysis measures how strongly that known starting condition lined up with the next period's market change, then inspects the link by ranked groups. A quartile sort splits starting conditions into four ranked groups and compares the average later return of each group.

When starting earnings yield was sorted into quartiles, each higher valuation-yield group was followed by a larger average 12-month index change, and the correlation between starting earnings yield and the next 12-month change was 21.8.

The lowest earnings-yield quartile, about 2.18 percent to 5.31 percent, was followed by a 4.50 percent average 12-month change, versus 12.54 percent after the highest quartile, about 9.04 percent to 16.50 percent.

Low starting valuation did not imply an automatic later decline: below-average earnings yields lined up with below-average but still positive subsequent 12-month changes.

The 10-year horizon still sorted by starting yield

Over a 10-year horizon, only starting earnings yields in the top three quartiles were associated with average total returns including dividends of 10 percent or more per year. Those quartiles began at 5.61 percent or higher, a price/earnings ratio of 17.8 or lower.

The highest quartile, 9.71 percent or more, saw about 16.6 percent a year.

Add the bond and bill backdrop

Intermarket analysis relates equity outcomes to conditions in other markets, especially bond and bill yields, so a stock trade is read against the rate backdrop.

Year-over-year declines in the 30-year Treasury yield lined up with stronger next-12-month index changes: the most-falling yield quartile averaged 12.14 percent versus 6.17 percent after the most-rising quartile, with a correlation of about -15.3 percent.

The same stepwise pattern appeared at the short end: when the 90-day bill yield was up more than about 28 percent year over year, the next 12-month index change averaged 4.80 percent, versus 13.28 percent when that yield was falling sharply, and the short-rate correlation, -21.2 percent, was stronger than the long-rate link.

Keep calendar markers on the same map

Seasonality analysis sorts later market outcomes by calendar markers such as month of year or the presidential election cycle.

Calendar context was treated as part of the same regime map: month of year and the presidential election cycle were listed among the historical factors that continued to sort later market outcomes, alongside earnings, dividend yield, and interest rates.

Evaluation asks whether those historically observed cross-market and seasonal associations actually sorted subsequent returns into distinct buckets. In this sample the quartile steps and the signed correlations show that they did for starting earnings yield and for rate changes.

Editorial reading: the stack does not turn any bucket into a single-path forecast. It asks whether the known-at-the-time mix of valuation, rates, and calendar state sits in a historically richer or thinner part of the ordinary 12-month map.