MA for Long vs Short ETFs

RyanInv

We use 200-day m.a. crossover as one of our signals. It's frustrating that the cross-overs don't occur on the same day/level for long index ETFs as they do for inverse ETFs. For example, look at pairs SPY/SH ad QQQ/PSQ. Can someone explain this?

markd

I'm no expert on inverse ETFs, so grain of salt.

They are different instruments that trade at different prices and for different purposes, so their averages will not necessarily be as mathematically in sync as you might expect.

Instruments that trade at different prices will have different (but usually at least congruent, if they follow the same instrument) per cent changes in values. The higher priced instruments will have finer per cent changes day to day vs. the lower priced instruments (e.g. a 1 point change on a scale of 10 is a 10% move, but 1% move on a scale of 100). Over time, even if small, the differences accumulate and could affect the cumulative value (the average) enough to change the day of the crossover.

Also, although the inverses are meant to mirror the longs, they don't exactly. The inverse may move more or less than the long. Possibly this is because the purposes of the funds are different and therefore the nature of the players in each are different. The long instruments appeal to buy and hold players, who are likely to be less active (except maybe in panic markets) whereas the inverses appeal to hedgers and short term players, who are likely to be more active (i.e. run beyond the nominal value of the index). So while arbitrage should narrow the excesses of inverses, it probably doesn't do so completely, so inverse prices would tend to move more independently of the indexes.