Which calendar clock changes a gold-versus-equity rotation test

What stays fixed

This archive evaluation treats a two-asset gold-versus-equity ranking as a fixed rotation rule and grades only the decision clock. The decision clock is the date on which the ranking or seasonal filter is observed and the next holding is booked.

Rank rotation, as used here, is a two-name procedure that holds only the higher-ranked asset for the coming interval and stays out of the lower-ranked name. The momentum-strategy form of that switch assigns the next holding interval to the asset that just outpaced its pair and uses that recent ranking as the entry, exit, and abstention signal.

Seasonal-trading in this setting is a calendar-gated allocation that lets a year-end close, a six-month mark, the first five sessions, or January decide the rest-of-period holding.

How the yearly clock is booked

A yearly two-asset rule holds, for the coming calendar year, whichever of a U.S. equity exchange-traded fund and a gold exchange-traded fund had the higher return in the year just ended.

Year-over-year returns are measured from the last December close to the next last December close on split- and dividend-adjusted prices. The evaluation assumes those two full-year returns can be estimated shortly before year-end and that the winning fund is bought at the final December close. It states that buying at the next-year open would not change the tested path.

The six-month restatement

The same winner-takes-the-next-interval ranking was restated on a six-month clock. Yearly figures were recovered by semiannual-annualization: the yearly mean as one plus the semiannual average, squared, minus one, and yearly volatility as the semiannual standard deviation times the square root of two.

Among the variations tested, the six-month restatement of the gold-versus-equity ranking was the procedure identified as stronger than the original yearly version.

First-five-days and January gates

A first-five-days seasonal gate holds the equity fund for the rest of the year if those five days are positive and otherwise switches into the gold fund at the fifth-day close.

A January gate waits until the January close, then holds the equity fund for the rest of the year after a positive January and the gold fund after a negative January.

In the 2005-2019 window, the first-five-days gate showed a higher average annual return and a lower standard deviation than the equity fund alone, while the January gate showed a lower average annual return than the equity fund alone.

A pair swap inside the yearly ranking

Substituting an emerging-market equity fund for the gold fund inside the same yearly ranking produced, over 2005-2019, a lower average annual return and a higher standard deviation than holding the U.S. equity fund alone.