2007 saw profitable closure of the LONG opened in 2006 and also 2 further profitably
closed LONGs. No losing trades.
During this year we turned $1,379,391 into $1,639,912. Please scroll down for links
to all weekly real-time reports.
We entered 2007 with an open
LONG position which was already showing a substantial profit (see the 2006 Performance Summary for details).
As the market rose in the first
weeks of the year our model began to turn increasingly neutral, dancing just above our liquidation point. We responded by trailing a stop upwards to protect our open profits.
This effectively gave us a two-way
bet; either the market’s momentum would continue to carry it forward and increase our profits or the trailing stop would
be hit.
In
the end, the latter occurred on 28th February 2007, closing our LONG at 2400 for a profit of 10.5% on the trade:
http://www.market-timing-signals.com/SQ Bulletin
2Mar07.pdf
Indeed, in the days that followed,
the market fell substantially and rapidly and there was a big spike in pessimism picked up by our model. Hence a new LONG signal was given out almost immediately as also signalled in the report of 2nd
March:
http://www.market-timing-signals.com/SQ Bulletin
2Mar07.pdf
This signal scaled us into yet
another full-blooded LONG position with total confidence (at an average NASDAQ level of 2345 deploying all our capital and
entered on 5th March 2007). Confirmation of the trade was given here:
http://www.market-timing-signals.com/SQ Bulletin
9Mar07.pdf
Indeed in March our model continued
to sniff out excessive bearishness as the model scaled a peak of 81% bullish probability by 16th March even after
we had already entered our trade with full capital.
This gave us a very high level
of conviction in the position, as evidenced for example by our commentary in the 16th March report: “The
speed of the recent sentiment shift makes it likely that our expected rally may be quite forceful when it arrives.” The full report is here:
http://www.market-timing-signals.com/SQ Bulletin
16Mar07.pdf
In fact, the market never really
went significantly below the level at which we entered the position and this was a textbook model trade.
There
was indeed a strong rally as many shell-shocked participants sat fearfully on the sidelines (a fact that had
been picked up by the model).
April was a particularly good
month as our fully leveraged position(committed with our full capital) bore fruit, just as the model had foreseen.
The
model then instructed us to be flat on 27th April:
http://www.market-timing-signals.com/SQ Bulletin
27Apr07.pdf
Hence we inserted both
a tight trailing stop and limit order to close the LONG as soon and as profitably as possible.
We were filled on 30th
April by the trailing stop at 2545, bagging us a full 200 points of profit (8.5% on the trade) in under 2 months. This was our 7th successive winning trade and is confirmed here:
http://www.market-timing-signals.com/SQ Bulletin
4May07.pdf
Over the summer we remained
in cash as the market rose but then fell back sharply in late July and early August.
The
model then gave us a further signal to go LONG at this time, so we set up limit orders in anticipation of picking up some
value for money fills (report of 10th August):
http://www.market-timing-signals.com/SQ Bulletin
10Aug07.pdf
These orders were filled for
most of our capital on 14th and 15th August at an average level of 2484. The report as of 17th
August gives the details here:
http://www.market-timing-signals.com/SQ Bulletin
17Aug07.pdf
At
this time, SQ peaked at 77% suggesting a high probability that the market would rise rapidly. As the chart on the Home page confirms (look for signal number 8 with the arrow), our model provided a superb entry level.
Many were baffled by the market’s
strong recovery in the second half of August though we were not: while August was traumatic for many hedge funds, our
strategy was one of the few to show a good profit over the month, having nailed the trough to go LONG.
September was equally profitable
for our open position as we gained hundreds of NASDAQ points above the favourable entry point.
On 5th October, the
model signalled that we should exit the trade and we therefore set up our usual limit/stop combination order:
http://www.market-timing-signals.com/SQ Bulletin
5Oct07.pdf
This time the limit order was
filled at 2799 (versus our average entry at 2484), giving us a nice profit of 12.7% on the trade:
http://www.market-timing-signals.com/SQ Bulletin
12Oct07.pdf
This second LONG trade of the
year was wonderfully-timed in that we were never underwater.
Although we remained in cash
for the rest of the year, the combination of two quick LONG trades netted us an
annual performance well ahead of most long-only and hedge fund strategies, while avoiding the summer volatility.
Key learnings for 2007
The 2 LONG trades entered this
year were textbook trades in that we entered rapidly and saw the market move in our favour almost immediately.
The trading strategy behind
the model still has certain glitches to be addressed; what cannot be disputed however, is the ContraQuant model’s accuracy
in picking its entry and exit points.
The one area where we could
have improved in 2007 would have been to keep a slightly tighter stop on the trade from late 2006 that was closed in late
February.
But
apart from that minor glitch, 2007 was the best year to date for the model’s performance.
