Daily Follow-Through in Up vs Down Trending Markets

30Jul08

As discussed previously, by reader request I’m taking some old posts and showing how the results differed in up versus down trending markets. This is a follow up to Evolving Markets and Dynamic Systems: Daily Follow-Through.

In this post, I’m going to be looking at follow-through, or put another way, if the market is up today, how likely is it to be up tomorrow, and if it’s down today, how likely is it to be down tomorrow?

I’ve reprinted the graph from my first post in which I showed that for most of the last 60 years, the S&P 500 has exhibited follow-through; up days tended to follow up days and vice-versa. However, since the mid 1970’s, this difference has been eroding and since the new millennium, it has flipped to contrarian. Up days now tend to be followed by down days, and vice-versa.

Here are those same results in up versus down trending markets for the S&P 500 from 1960 to present. I’m defining an up/downtrend as any time the market closes above/below its 200-day SMA. Note that this graph uses a 10-year moving average of next-day returns (rather than our original 5-years).

Through most of the market’s history, a downtrend has magnified the results of the earlier report and increased the degree of follow-through. Closes up in a downtrend were followed by higher returns than in an uptrend, and conversely, closes down were followed by even lower returns.

However, since the 1990’s this “magnification-effect” has been decreasing to zero. At the moment, not only is daily follow-through slightly contrarian (closes up are followed by closes down and vice-versa), but this observation is not affected by the broader trend.

Happy Trading,
ms

To stay up to date with what’s happening at the MarketSci Blog, we recommend subscribing to our RSS Feed.

Geek notes: (1) returns have been normalized by subtracting the average return of all days in each observation period to remove the influence of bull vs bear markets, (2), daily returns have been capped at +/- 5 standard deviations to mitigate the influence of a single large day, (3) averages are geometric.