It is not a secret that the stock market tends to perform better during the end of month/beginning of month and middle of month periods than during the rest of the month. What is less known is that certain sectors tend to perform even better than the overall market during these time periods.
(See also Where NOT to Invest, Er, Soon (Part 1))
Jay’s Seasonal Health Care Strategy
As always, what follows is not a “recommended trading strategy”. It is presented solely as “food for thought” (or perhaps I should say “meds for thought”). Anyway, for illustrative purposes we will adopt an aggressive trading strategy as follows:
*We will use Health Care UltraSector ProFund (ticker HCPIX, which tracks the Dow Jones U.S. Health Care index using leverage of 1.5-to-1; so I the index rises 1% today the fund should rise 1.5%);
We will hold HCPIX ONLY on:
*The last 4 trading days of the month
*The first 2 trading days of the month
*Trading days numbers 9, 10, 11 and 12 each month
The rest of the time we will hold cash; For the purposes of this test, no interest is assumed to be earned while out of HCPIX
Figure 1 displays the hypothetical growth of equity using the approach.
Figure 1 – Growth of $1,000 using Jay’s Seasonal Health Care Strategy with ticker HCPIX (6/10/2000-5/13/2016)
For the record, $1,000 starting on 6/19/2000 grew to $20,995, or +1,999%.
Figure 2 displays the year-by-year results.
Figure 2 – Jay’s Seasonal Health Care Strategy Year-by-Year (6/19/00-5/13/16)
One question that invariably arises when talking about a specific seasonal approach such as this is, “What do I miss by being out of HCPIX the rest of the time?” Based on my own thorough analysis I believe the answer to that question is “Pain and suffering”.
Figure 3 answers the question of what would have happened to an initial $1,000 invested in HCPIX only during “all other days” besides the ones listed earlier.Figure 3 – Growth of $1,000 in HCPIX during all other trading days (6/10/2000-5/13/2016)
For the record, starting on 6/19/2000 (when HCPIX started trading):
*$1,000 invested only during the seasonally favorable periods grew +1,999% (to +$20,995)
*$1,000 invested only during all other days shrank -89% (to $111)
These are the kinds of disparities that we “quantitative types” refer to as “statistically significant”.
Jay Kaeppel
is there an ETF that matches HCPIX including the 1.5X factor?
Paul, yes but….there is CURE which is 3x and RXL which is 2x. However, volume for CURE is OK but relatively low (140K shares) and RXL is very low (12K shares). The most heavily traded non-leveraged is XLV (11 million+ shares). Jay
Is health care the single best profunds sector fund for this strategy? What are the next 2 runners up, in case once wanted to diversify the strategy a bit?
What are these numbers & if how do I use it
*Trading days numbers 9, 10, 11 and 12 each month
You only count the days of the month on which actual trading takes place. So if the 1st of the month is on a Monday for example – and assuming no holidays – then Monday through Friday is trading days #1 through 5. Then Monday through Friday of next week is trading days 6 through 10. Hope that helps, Jay
Hey Jay,
In considering running multiple models like above, semi conductors and your umds system from your book, what would be the best way to study correlation. Put monthly or weekly returns together and run correl study. Or simply run all 3 at a smaller size for diversification.
Would also be really interesting to hear how you allocate out to your various seasonal edges?
Cheers Ben
Hen, I like diversifying among various strategies. The keys are finding things that (hopefully) offset one another’s down periods to create as smooth of an overall equity curve as possible. A smooth equity curve is what keeps people disciplined. Correlation studies are useful but do remember that correlations can change over time so those things are never set in stone. I like putting equity curves for various systems/methods together and see what the fluctuations look like (average x-month Return, % of x-month returns that are positive, typical drawdowns, etc.)
Hope that helps.
Jay