Post New

This is a cross post from the Index Investing, which at times it is philosophically a close sister to this board.

https://www.shrewdm.com/MB?pid=643226347

- Manlobbi

Annual returns for the gtr1 screen {zgYPS} are close to the ETF YPS:

year  zgYPS  YPS  RSP  SPY
2018    -9   -10  -8   -5
2019    28   28   29   31
2020    12    8   13   18
2021    33   31   29   29
2022   -12   -13  -12  -18

Backtest results are close to SPY, RSP, and S&P 1500 equal weight (with no friction, quarterly rebalancing).

from 19570301 to 20230310:

Screen  CAGR  SAWR  GSD  MDD  Sharpe  AT   Depth  weight
zgYPS   12.5  8.4   24   -65   0.46   0.6   500   inverse
 gSPY   10.3  6.2   17   -55   0.44   0.3   499   mkt cap
zgRSP   12.0  8.1   19   -60   0.50   0.3   500    equal
SP1500  12.2  8.5   20   -60   0.49   0.3   930    equal

from 19870302 to 20230310:

Screen  CAGR  SAWR  GSD  MDD  Sharpe  AT   Depth  weight
zgYPS   10.7  7.4   26   -65   0.43   0.6   501   inverse
 gSPY   9.9   5.4   18   -55   0.49   0.3   500   mkt cap
zgRSP   11.1  7.9   20   -60   0.52   0.4   501    equal
SP1500  11.4  8.5   22   -60   0.50   0.4  1287    equal

from 20130308 to 20230310:

Screen  CAGR  weight
zgYPS   11.2  inverse
 gSPY   11.7  mkt cap
zgRSP   11.2   equal
SP1500  10.6   equal

Define {zgYPS}
step0: [S&P 500 Member; lag=1 days] == 1
Holding period = 63 mkt days
Weight liquid positions by [inverseMktCap] every holding period
No friction
https://gtr1.net/2013/?~zgYPS:h63::sp500.a:et1:Mkt...

There was high start date variability in 2020, with 2020 annual results ranging from 7 to 25 for the 63 different start dates. (Standard deviation of CAGR was 3.66. YSP results were about 1.2 standard deviations from the {zgYSP} mean. 1.2 = (12.14 - 7.70) / 3.66 ) Expect high tracking error in volatile years. For example, 2009 annual returns ranged from 55 to 105, with a standard deviation of 9, depending on the start date.

Manlobbi, you wrote "As a stock declines, an equal weight index buys more of it, but only subtly."

Why would such an index buy more? Isn't the position fixed at, say, 1/500th of the index value?

DB2

Manlobbi, you wrote "As a stock declines, an equal weight index buys more of it, but only subtly." Why would such an index buy more? Isn't the position fixed at, say, 1/500th of the index value?

The target allocation of an equal weight index is frequently rebalanced to equal weight. Take QQQE for example, with the target allocation being a 1% holding of each of the largest 100 stocks in the NASDAQ. If one firm doubles in price to then become a 2% position, then it needs to return to 1% so the index sells half the holding to return it to 1%. Conversely if a firm halves in value to 0.5% of the index then a 0.5% position is purchased to return it to 1%.

The equal weight index forces systematic selling of stocks rising in quotes, and buying of stocks that fall in quotes.

Much, but importantly not all, of this rising quotes is associated with rising value. As a result there is a mild effect of the equal eight index selling overpriced positions and purchasing more underpriced positions. Call this the 'trading advantage'.

It also had the disadvantage of not allowing considerable "winners to run". That has worked well for cap weight indexes the last 5 years, but was not always such a large advantage.

The second advantage of equal eight indexes is that they avoid avoid a bias of exposure in firms that already have market dominance and may have less room to grow. Call this the 'economic advantage'. In the long past history going back a century, the largest firms performed worse than the medium sized firms because once you have 'made it', you could no longer have a 100x return from there except through the effect of dividends over very many decades. However the last ten years have been a historical aberration, with the largest firms continuing to have superior returns, thus the ordinary cap weighted Nasdaq QQQ recently outperformed the equal weight Nasdaq QQQE. That may continue for a while but longer term I would rather bet on 100 years of history than 10.

- Manlobbi

Manlobbi, you wrote "As a stock declines, an equal weight index buys more of it, but only subtly."

Why would such an index buy more? Isn't the position fixed at, say, 1/500th of the index value?

When the ETF is rebalanced there would of course be some selling and buying maybe that is what he means by "subtly".

GD_

Post New

US Policy FAQ

---

Contact Shrewd'm
Contact the developer of these message boards.