|
PureBytes Links
Trading Reference Links
|
Gitanshu,
Great work. I've been thinking along those lines myself and I'm glad
someone has beat me to it and has already done all the leg work. I'm going
to think about how to incorporate your findings in the first part of my
trading strategy which is to pick the right sectors. In the past, I've just
waited for a sector to begin to trend strong and then jump on. If the
seasonality holds true, it will have the great advantage of being able to
jump on early at lower prices before the trend starts. Keep up the good
work.
Once again, thanks for sharing.
JimG
----- Original Message -----
From: "Gitanshu Buch" <OnWingsOfEagles@xxxxxxxxxxxxx>
To: <metastock@xxxxxxxxxxxxx>
Cc: <realtraders@xxxxxxxxxxxxxxx>
Sent: Monday, December 27, 1999 7:46 PM
Subject: Heat Map explained
> Before this gets overblown into some needless controversy - I am
explaining
> what this spreadsheet is all about right here.
>
> The reasons this spreadsheet was not uploaded to RT/Metastock before were
>
> - simply because I thought nobody here would be interested, hence it was
> shared with a few close friends before the public release @ Microsoft
> MoneyCentral.
>
> - I felt kinda embarrassed tooting my own horn.
>
> It seems to me that the latter was a needless concern. Grandma phrases
like
> God helps those who help themselves etc etc come to mind.
>
> Here is how this started:
>
> Jon Markman, an author (ex LA Times, now Microsoft, and responsible for a
> bestseller entitled Online Investing) wrote an article publishing some of
> his research on stock-specific seasonality.
>
> This research was astounding to me - in its simplicity, and verifiability.
>
> Here is that article:
> http://moneycentral.msn.com/articles/invest/models/4804.asp
>
> Independently, and unrelated to the above, I have observed and traded
> seasonal patterns successfully in the energy derivatives complex. My work
> has been visual converted to Excel, clumsy, but gratifying since it has
been
> profitable.
>
> I got to know him, and we started exchanging ideas. One thing led to the
> other, I did the index research, put it on a user-friendly spreadsheet,
and
> shared it with Jon. He got excited, and wrote it up in an article at their
> website.
>
> This started out as a discovery. It is aquiring a life of its own -
perhaps
> the way of the world in internet time - for that has since resulted in
about
> 200 emails in the past 3 weeks from strangers seeking investment advice, a
> book offer from someone in London, and assorted other flora and fauna.
>
> Now.
>
> Do I care it gets mass-distributed? Heck, no. This wouldn't be in public
> domain otherwise.
>
> The spreadsheet focusses on indices that have liquid, tradeable options
> because that sphere is where I trade the funds I manage and that is where
I
> can find direct measurable portfolio impact.
>
> However, that work has now expanded to include Fidelity Select Sector
Funds,
> and I am happy to share that work here if anybody is interested. Fund data
> is currently inaccurately adjusted for dividends/distributions and being
> cross-checked - but the trends exist and the trends are tradeable and I
> intend to trade them with real money come Jan 2000.
>
> The following is the text of what was sent to Jon before he wrote it up in
> his article - which is at this url.
> http://moneycentral.msn.com/articles/invest/models/4950.asp
>
> Microsoft owns all the copyrights for all urls referenced in this space.
>
> Finally, and once again - I mean no disrespect towards Mark the person -
my
> previous email was towards the act - which looked and felt like plagarism
in
> the crude way the email landed in my face - I believe I asked for basic
> decency out of sheer indignation.
>
> ====
> The spreadsheet as explained:
>
> Sector picking (through Sector Mutual Funds or their tracking sector
> stocks) - beat S&P 500 Indexing by a factor of 10 to 1 since 1994 using NO
> LEVERAGE. The results are remarkably better if individual stocks are
chosen.
>
> Methodology:
>
> PRQs were tested across 150 companies in different industry sectors, just
to
> establish the validity of the concept.
>
> Research then extended to 31 industry sector indices, for as far back as
> sector index data could be found.
>
> This spreadsheet is the result of the industry sector research. It
provides
> a "Heat Map" for sector rotation that seems to evolve in the market.
>
> The Heat Map is set about as follows:
>
> Each column has data for a single index across the 12 months.
>
> Each row has data for a single month, across the page index.
>
> Green cells are "Best Months to Buy these sectors"
>
> Red cells are "Worst Months to Buy these sectors; conversely, best months
to
> Short them".
>
> Blue cells are Best of Breed for the month, across all sectors researched
so
> far.
>
> A summary of the best 3 sector returns for each month is listed in the
> initial 3 gray columns, and the Benchmark index is S&P 500.
>
> To use this to its desired initial screening potential or even as a
> standalone strategy, the reader would:
>
> Buy the best performing index for Dec at the Close on the last trading day
> of Nov.
>
> Rotate into the next month in similar fashion on the last trading day in
> Dec.
>
> Assuming no leverage or transaction costs, the trader outperforms SnP 500
> Indexing by the average factor listed in Row 21. (An outperform reading of
1
> equals the benchmark, of 0 under-performs, and of 10 beats the benchmark
10
> to 1).
>
> The trader would further research the index components' chart patterns and
> PRQs for individual stock selection. This component list (and where given,
> weight within the index) is hyperlinked with the Index name in Row 2. Just
> have an internet connection handy.
>
> The work-in-progress is as follows:
>
> I am PRQ'ing sector components, and then doing the Heat Map for each stock
> in that sector at Month Level.
>
> The Month level will then be imploded to explore the best weeks in the
month
> to hold that stock, or its options. Given that this data captures an
entire
> month while options expire on the third Friday of each month, there will
be
> some fine-tuning. But the core concept seems to be ready to be set in
> stone - the findings are really exciting:
>
> Some Findings:
>
> A/ Forest and Paper Products is the best performing sector in April of
every
> year. FPP appreciates better than SPX by a factor of 10. Better than
> Internets, Semis, Banks, you name it. This is consistent with Jon's
original
> PRQ article and his experience with Dow 30 component Cyclicals.
>
> B/ Semiconductors have a tendency to outperform the rest of the world and
> their own annual buy-and-hold results in the January, April, and July of
> each year - notably, these are the first months in every quarter.
>
> C/ Networking stocks are best bought in April and held through July. This
> beats the SPX by a factor of 9 to 1 during the same period.
>
> D/ While September is billed to be the worst month for the market (true),
> the sectors studied actually end up losing the most of their value in
> August. This leads me to believe that the September under-performance
> catches up with the not-too-favored stocks that make up the general index,
> while the real damage in the market's YTD faves happens in August.
>
> E/ Contrary to current popular belief, Oil prices, Oil sector stocks and
> Airline stocks can, and do rally together - the spreadsheet will show that
> these sectors simultaneously outperform the rest of the world (including
> Internet stocks) in March of each year.
>
> F/ Airlines again outperform the universe in Oct and Nov each year, but
Oil
> sector stocks curiously decline in this period.
>
> G/ While on Oil: This is the classic Anti-January effect sector. The
sector
> plunges >5% on average every year, (the worst performer of the universe)
> making it a prime candidate for shorting during the month.
>
> H/ The best January Effect halo exists over Brokers, Semiconductors, and
> Internets. While other sectors do find investor enthusiasm, it appears
that
> following the crowd while ignoring these 3 sectors obligate you to
> mediocrity, for you give up >4 to 1 returns in a generally strong market.
>
> I/ There is something to be said about being defensive. Drug related
> sectors, for example, typically outperform the benchmark during the
market's
> weak period in Aug-Sep. But within the drug sector, the more momentum
driven
> Biotech sector is better owned than the Mercks of the world.
>
> Next steps for me:
>
> Sector-Component Level seasonality
> Week-of-month seasonality at Sector and Component Level.
> Trading Instrument choice and Trading Plan finalization.
> Since I now have a general heat map, I will start with sectors that enjoy
> their best seasonality in Dec and Jan. If you are interested, I will send
> these spreadsheets over.
>
> Weaknesses/Shortcomings noticed so far:
>
> a. Data inconsistency: Index measurements start from different dates in
> history, a little skew may exist.
> b Multiple index participation: Lots of stocks appear in multiple index
> standings.
> c. Unknown reasons for some stock exclusions: For eg, the SnP Bank Index
> (BIX) does not have Citibank as a component. Go figure.
> d. Bad data on MSN/CSI database. Can't use it unless manually washed.
> e. From a trader's perspective, given today's ATR, a lot of these numbers
> seem miniscule - a whole month's performance can be reached within a week
or
> less.
>
> Comments welcome.
>
> Gitanshu
>
>
|