The Golden Cross and DJT Confirmation

From MarketSci Blog by marketsci

A lot of chatter about the stock market’s upcoming golden cross has focused on whether the Dow Jones Transportation Index (DJT) is confirming a bullish trend.

Unfamiliar with the golden cross? Read our take on the cross, about its two flavors (SMA and EMA) and crosses in up vs downtrending markets.

As the theory goes, to qualify as bullish, the DJT (an index of U.S. companies that move stuff around like airlines, railroads, ocean freight, etc) must also confirm the market’s golden cross. I’m not a Dow Theory proponent, but if I remember correctly, this concept originated there.

In this post, we’ll look at how off-the-shelf golden crossovers, versus those confirmed by the DJT, have performed historically.

[logarithmically-scaled]

The chart above shows the results of two strategies trading the Dow Jones Industrial Average (DJIA) from 1931 (blue) to present. The first (red) is the straight version, going long at today’s close if the 50-day SMA crossed above the 200-day today. The second (green) is looking for DJT confirmation – it will only go long if both the first condition is met AND the 50-day SMA of the DJT index is trading above its 200-day.

Geek notes: these results are frictionless (i.e. do not account for transaction costs or slippage), and I’ve included a return on cash when not invested of half the nearest 13-week Treasury bill.

And for the number lovers…

Requiring DJT confirmation, improved performance over the life of the test, but only slightly. The strategy has gone through long periods where it hasn’t really helped or hurt.

What is more impressive to me is that the confirmation strategy maintained performance, while drastically reducing exposure (time invested in the market).

I’ve written about this before: accomplishing the same end goal with less exposure to the market is an inherently good thing because it reduces the risk of getting caught looking the wrong way on a massive unpredictable black-swan’ish day (a’la Oct. 1987 or Sep. 11, 2001).

Do I think DJT confirmation, as I’ve defined it here, is a huge step up in the strategy? No. Do I think it’s a nice little tweak that should at least be in the back of the mind for trend-following types? Sure.

Last comment: I’m not a fan of trading the DJ Industrial Average. I used it here because it was a nice match for the DJT, but my personal opinion is that it’s too driven by too few names (in other words, it’s not sufficiently diluted) to be easily traded.

I personally would much rather trade something like the S&P 500, so in a follow up post, I’ll rerun this same golden crossover test on the S&P 500, but still using the DJT for confirmation. More to follow.