No. of Recommendations: 10
This is a follow-up from a screen posted by musselmant
musselmant replies on study of top 10 52wk mom stocks in Nasdaq100Variations on 52 wk momentum screen in Nasdaq100
Part 1 of 2
Backtest results can depend heavily on the choice of start date. The
problem with Nasdaq is the dot-com boom from 1999 to 2000 where it
doubled, and then the dot-com crash from 2000 to mid 2001 which was a
75% loss.
A start in 1995 or 2003 or 2023 will show outstanding results. A
start in 2000 or 2022 will show not so good results.
Here are the results from the turn of the century, January 2000.
GTR1 results report has one statistical problem. It reports the
results for every start day in the trading cycle, and the average of
these. But the average can get skewed by one outlier high.
Reporting the median would be better.
When evaluating backtests it is probably safer to assume you will get
the _lowest_ of the cycles rather than the average or median. Better
for surprises to be happy rather than sad.
As reported below, the various slightly different parameters give
these CAGR results:
15.2% Original
14.0% Do not rebalance
14.2% Do not rebalance and do not use BCC timing
13.8% Use RS instead of TR (no timing)
14.4% Add HTD 12 (no timing)
15.5% Change RS to "price in 52 week range" (no timing) HTD 12
8.1% S&P 500 total return with 19.4% Volatility
8.2% QQQ total return with 26.9% Volatility
Original GTR1 screen:
gtr1.net: GTR1 Backtester(Step 1: "aprc(1) > 0" means price is greater than zero. Not sure why
this is here. Maybe to filter out data errors?)
Changing the starting date. 2000-01-03 to 2025-11-28
gtr1.net: GTR1 Backtester
Rows are: CAGR, GSD, AT
Rounded to 1 decimal digit for readability.
The first field is the average, the remaining fields are starting day cycles.
15.2 14.6 14.3 15.0 15.9 14.9 15.2 14.7 14.6 15.0 15.6 15.2 15.4 15.9 15.4 15.9 16.0 16.2 14.9 14.9 15.0 15.0
26.8 26.6 26.7 26.8 26.8 26.6 26.6 26.8 26.7 27.0 26.9 27.2 27.1 26.7 27.0 26.9 26.6 27.0 27.0 26.5 26.4 26.8
4.4 4.4 4.3 4.4 4.4 4.4 4.4 4.3 4.3 4.4 4.4 4.4 4.5 4.4 4.4 4.4 4.4 4.4 4.4 4.4 4.3 4.4
When evaluating backtests it is probably safer to assume you will get
the _lowest_ of the cycles rather than the average or median. Better
for surprises to be happy instead of sad.
Reporting median and sorted cycle returns (by CAGR).
15.0 14.3 14.6 14.6 14.7 14.9 14.9 14.9 15.0 15.0 15.0 15.0 15.2 15.2 15.4 15.4 15.6 15.9 15.9 15.9 16.0 16.2
26.8 26.7 26.6 26.7 26.8 26.5 27.0 26.6 26.8 26.4 27.0 26.8 26.6 27.2 27.1 27.0 26.9 26.8 26.7 26.9 26.6 27.0
4.4 4.3 4.4 4.3 4.3 4.4 4.4 4.4 4.4 4.3 4.4 4.4 4.4 4.4 4.5 4.4 4.4 4.4 4.4 4.4 4.4 4.4
Note that the CAGRs range from 14.3% to 16.2%. Quite a jump. Better
to expect 14% and get 16% than the reverse. </editorial>
Same rules except 0 for never rebalance.
gtr1.net: GTR1 Backtester
CAGR, GSD, AT
14.0 12.9 12.8 13.9 14.7 13.9 13.9 13.5 14.0 15.0 15.5 15.1 14.4 14.1 13.0 13.0 14.1 14.6 14.2 14.0 13.8 13.3
25.9 26.2 25.9 25.9 26.1 25.9 25.8 25.7 25.7 25.9 25.8 26.0 25.9 25.9 26.2 26.1 26.0 26.1 26.1 25.6 25.8 26.3
3.0 3.0 3.0 3.0 3.1 3.1 3.1 2.9 2.9 3.0 2.9 2.9 3.0 2.9 3.0 2.9 3.0 2.9 2.8 2.8 2.8 2.9
Same rules except 0 for never rebalance, and eliminate the BCC step.
gtr1.net: GTR1 Backtester
14.2 13.5 12.6 13.6 13.9 14.7 13.8 13.2 14.8 14.6 14.8 14.6 13.7 14.7 13.2 14.5 14.8 16.2 14.4 13.1 15.5 14.1
34.4 34.1 34.9 35.3 35.0 34.2 34.3 34.8 34.1 34.4 34.8 34.4 35.2 33.8 34.6 33.9 34.3 34.1 34.3 34.4 33.9 34.4
3.1 3.1 3.0 3.0 3.1 3.1 3.1 3.0 3.0 3.0 3.0 3.0 3.1 3.1 3.1 3.1 3.1 3.1 3.0 3.1 3.0 3.1
(Slightly better CAGR and much worse stdev.)
Change from tr (total return) to rrs (relative strength). Because rs
is readily available in databases but tr is harder to find.
gtr1.net: GTR1 Backtester
13.8 13.6 13.3 13.1 13.1 14.0 13.9 13.1 13.5 14.1 14.2 14.5 14.5 14.5 13.8 13.5 13.5 13.9 13.8 14.0 13.1 14.6
34.6 34.4 35.0 34.8 34.8 34.9 34.7 34.7 35.1 35.0 34.8 34.8 34.9 34.4 34.7 34.2 34.1 34.3 34.4 34.0 34.1 34.3
2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.0 2.1 2.1 2.1
Adding "HTD 12"
gtr1.net: GTR1 Backtester
14.4 13.4 13.7 13.6 14.4 15.4 13.5 13.1 13.5 13.9 15.1 15.8 15.7 15.5 15.6 13.9 13.4 14.8 15.2 15.2 14.7 13.9
35.0 34.9 35.0 35.5 35.7 35.5 35.2 35.2 35.6 35.1 35.4 35.4 35.3 35.1 35.3 34.4 33.9 33.8 33.9 34.4 34.5 35.1
1.8 1.9 1.8 1.8 1.8 1.7 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.7 1.7 1.8 1.8 1.9
-------------------------------------------------
A Jim idea, ranked by current price vs. 52 week range. Also HTD 12.
gtr1.net: GTR1 Backtester
15.5 17.3 15.7 14.0 15.4 13.8 14.6 15.0 14.6 16.3 14.3 14.4 13.2 14.4 14.0 14.3 16.4 16.7 17.8 16.7 18.2 18.3
31.8 31.8 31.6 31.4 31.4 32.0 31.5 31.7 31.7 32.3 32.4 32.0 32.8 31.6 31.8 31.8 32.2 31.6 31.5 32.0 31.4 31.4
3.4 3.5 3.4 3.5 3.4 3.4 3.4 3.4 3.4 3.5 3.3 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.5
(HTD 15 has about the same CAGR & GSD but lower turnover.)
