Академический Документы
Профессиональный Документы
Культура Документы
FOREX WATCH 2009 p. 8 THE EURO: Back in the saddle or just a bounce? p. 14 INSIDE-DAY FADE: Simple setup proves resilient p. 20
CURRENCY TRENDS: The majors p. 26 SPOT CHECK: Dollar vs. Euro relative performance p. 32 FED CUTS and the fate of the buck p. 38
CONTENTS
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . .6
Advanced Strategies Let the trend be your friend: The majors . . . . . . . . . . . . . . . . . . . . . . .26
If currencies trend so much, why do trend followers usually have such blah performance? This and other questions are answered in this study of currency trends. By Howard L. Simons
continued on p. 4
FREE 25
$
www.fxcmmicro.com/ct
Offer expires 2/1/2009
Restrictions apply. Please visit the URL listed above for details. *NOTE: FXCM strives to provide traders with tight, competitive spreads; however, there may be instances when market conditions cause spreads to widen beyond the typical spreads displayed here. WARNING: Trading foreign exchange on margin carries a high level of risk and may not be suitable for all investors. Before deciding to trade foreign exchange, you should carefully consider your monetary objectives, level of experience, and risk appetite. The possibility exists that you could sustain a loss of some or all of your deposited funds, and therefore you should not speculate with capital that you cannot afford to lose. You should be aware of all the risks associated with foreign exchange trading and seek advice from an independent advisor if you have any doubts. Past returns are not indicative of future results.
CONTENTS
Global Economic Calendar . . . . . . . . . . . .41 Forex News Rate cuts fuel deflation fears . . . . . . . . . .38
Inflation continued to drop as the Fed cut its target lending rate to a historic low in December, amplifying the dollars mid-month drop against major currencies. Important dates for currency traders.
CONTRIBUTORS CONTRIBUTORS
Howard
Simons is president of
Rosewood Trading Inc. and a strategist for Bianco Research. He writes and speaks frequently on a wide range of economic and financial market issues.
Barbara Rockefeller (http://www.rts-forex.com) is an international economist with a focus on foreign exchange. She has worked as a forecaster, trader, and consultant at Citibank and other financial institutions, and currently publishes two daily reports on foreign exchange. Rockefeller is the author of Technical Analysis for Dummies (For Dummies, 2004), 24/7 Trading Around the Clock, Around the World (John Wiley & Sons, 2000), The Global Trader (John Wiley & Sons, 2001), and How to Invest
Volume 6, Issue 1. Currency Trader is published monthly by TechInfo, Inc., 161 N. Clark Street, Suite 4915, Chicago, IL 60601. Copyright 2009 TechInfo, Inc. All rights reserved. Information in this publication may not be stored or reproduced in any form without written permission from the publisher. The information in Currency Trader magazine is intended for educational purposes only. It is not meant to recommend, promote or in any way imply the effectiveness of any trading system, strategy or approach. Traders are advised to do their own research and testing to determine the validity of a trading idea. Trading and investing carry a high level of risk. Past performance does not guarantee future results.
Internationally, published in Japan in 1999. A book tentatively titled How to Trade FX is in the works. Rockefeller is on the board of directors of a large European hedge fund.
GLOBAL MARKETS
FIGURE 1 EURO RETRENCHMENT After the yen, the U.S. dollar jumped the most during the late-2008 market crisis.
Source: TradeStation
FIGURE 2 EURO/DOLLAR The dollar might struggle vs. the Euro, at least early in 2009.
mid the storm of financial and economic crises that dominated the year, many traders were happy to close the door on 2008. Last years action in the foreign exchange market was volatile, with fall and winter action driven by massive risk aversion and global position liquidation. The Japanese yen came out on top of the 2008 carnage, chalking up a 23-percent gain vs. the dollar. The main factor propelling the yen was a massive exit out of carry trades, which sparked doubledigit losses in many of the high-yielding currencies that were the on the long side of this popular trade. The Australian dollar (AUD) dropped 21 percent vs. the dollar through Dec. 30, the New Zealand dollar (NZD) declined 25 percent, the Brazilian real plunged 24 percent, the Mexican peso sank 22 percent and the South African rand (ZAR) plummeted 27 percent. The British pound (GBP), which posted a dismal 27.3 percent drop vs. the U.S. dollar as of Dec. 30, was the weakest of the most liquid currencies, according to John Rothfield, senior currency analyst at Bank of America. He says Britains paper suffered especially because its economy is so reliant on the financial and housing sectors. The most positive performers, Rothfield adds, were the low-yielding currencies pointing to the yen and the Swiss franc, which gained 7 percent vs. the dollar.
Dollar action
The U.S. dollar was the second-biggest beneficiary of the economic crisis, as funds flooded into the buck as a safe haven (Figure 1). The U.S. dollar ended the year with a modest 4.8-percent gain vs. the Euro, as August through November saw a sharp reversal of the bear market action that has dominated the greenback since 2002. The Euro/dollar (EUR/USD) plunged to $1.23 in November from the 1.6000 July high (Figure 2) a 38.2-percent retracement of the 2002-2008 bear market for the U.S. dollar. Into year-end, however, the U.S. dollar gave back some its gains, as the immediate financial panic subsided and the U.S. made a historic shift to a
January 2009 CURRENCY TRADER
Source: TradeStation
near-zero interest rate policy: On Dec. 16, the U.S. Federal Reserve slashed the Fed funds rate to a range from zero to 0.25 basis points, which detracts from the dollar from an interest-rate differential perspective. The Euro/dollar had climbed back above 1.4000 by year-end. Rothfield says U.S. fundamentals arent particularly good. He chalks up a large portion of the massive fall rally in the dollar to unusual market conditions and a huge shortage of the dollar globally, which triggered safe-haven buying in the greenback and U.S. Treasuries. We think in the early part of the year through the spring the U.S. dollar will struggle under the weight of U.S. quantitative easing, he says.
ments that suggest muted interest-rate easing. The ECB has built a tremendous amount of market credibility by not shocking the market, Woolfolk says. In addition to interest-rate differentials, growth differentials also favor a continuation of the recent Euro/pound trend. Bank of New York Mellon is forecasting a 2-percent decline for UK 2009 gross domestic product (GDP) vs. a 1-percent decline for the Eurozone. The recession has not hit central Europe as acutely as it has hit the UK, Woolfolk says. Woolfolk, for one, sees the potential for the uptrend to extend beyond the parity level into the new year until the ECB signals that it is prepared to adopt a zero interest-rate policy along with the U.S. and UK, he says. Finally, Woolfolk says recent trade data is also bullish for the Euro vs. the pound. The trade figures in the Eurozone are relatively stable and balanced, vs. a large trade deficit in the UK, he says. He says October 2008 data showed an EU trade surplus of 900 million vs. a deficit of 3.9 billion for the UK. On the downside, Woolfolk cites .9250 as key support. If we were to close below that level, it would take out this bullish uptrend, he says. Woolfolk does warn of the potential for buy the rumor, sell the fact action around the early January BOE meeting and says traders may get in early and take profits if the BOE does what they are expecting. But once the Euro/pound conquers the parity level in the weeks ahead, Woolfolk predicts forex traders will continue eyeing round numbers at 1.0100 and then 1.0200. Woolfolk advises traders to watch for indications the ECB is considering a zero interest-rate policy, which could prompt significant profit-taking.
Source: TradeStation
FIGURE 4 AUSSIE/CANADA Economic fundamentals in favor of the land down under make the Aussie/Canada pair a market to watch in coming months.
Aussie/Canada
Source: TradeStation Another cross with potential for movement is the Australian dollar/Canadian dollar (AUD/CAD) pair, according to Brian Dolan, chief currency stratescheduled to meet next on Feb. 3. Further rate cuts are gist at Forex.com. The pair was trading around 0.8500 at expected, with Dolan forecasting a bottom to the RBA easyear-end (Figure 4), and Dolan sees a possible move to ing cycle around 3.25-3.00 percent, which would still favor 0.9500 or parity. the Aussie dollar over the Canadian currency. Dolan believes a number of factors including interest-rate Dolan also interprets a more bullish commodity-export differentials, commodity exports, and regional growth picture for Australia. opportunities are tipping in favor of Australia in the new Both are considered commodity currencies, but the big year. difference is that Canada is oil reliant and I expect oil prices The Bank of Canadas (BOC) lending rate currently stands to remain weak, he says. at 1.5 percent; a 0.50-percent cut is expected at the Jan. 20 Australia is the worlds leading coal exporter and Dolan meeting. That compares to the 4.25-percent rate target cur- saw continued massive demand for coal from China. continued on p. 12 rently held by the Reserve Bank of Australia (RBA), which is
10
FIGURE 5 AUSSIE/U.S. DOLLAR Some analysts believe the Australian dollars sell-off was overdone and the currency is poised to rebound.
There is regional support for Australian commodity output, he says. Canada, Dolan notes, is closely tied with the U.S., which is in the midst of a recession, while Australia is close in proximity to the Asian region. Generally speaking, the Asian region seems to be faring best out of all the regions globally, which will support Australia, Dolan says. Dolan thinks the .8250 to .8400 zone is a good buying area for Aussie/Canada. Try to pick it up if we get some weakness near term, he says. Dolan says the strategy would be negated on a drop below .8000 if that were to occur. On the upside, he sees .9500-1.000 as an objective.
Playing it safe
There may be opportunity, but forex traders should be cautious in the new year as volatile conditions could reemerge at any time. People are looking for singles as opposed to doubles and home runs in the early part of the year because of the uncertainty on the global front, Rothfield says. People had been very defensive in the fourth quarter of the year and had to sell good stuff to offset losses in other areas. We will get some further unwind of the extreme risk aversion and see a temporary return of risk appetite and search for yields.
January 2009 CURRENCY TRADER
12
ON THE MONEY
he consensus forecast today for the Euro/dollar (EUR/USD) is a move up to 1.5000 and probably higherperhaps a test of the July 2008 high above 1.6000. But this forecast is based on few facts and questionable assumptions. All forecasts entail creating a scenario for the most likely outcome. Sometimes its all too easy to convince yourself a scenario is highly likely given certain facts and reasonable
assumptions. But in a crisis situation such as todays, not even facts are reliable and just about every assumption should be looked at with suspicious eyes. These are the times we question whether what we think we know is true, and wonder whether what we dont know is going to jump up and bite us on the nose. This is true of fundamentals and technicals alike. Figure 1 shows the EUR/USD on a weekly basis. The Euro uptrend from the 2000 low to the 2008 high is clear. FIGURE 1 BACK TO NORMAL The precipitous drop starting After the precipitous drop that started in October 2008 the EUR/USD is moving back in October 2008 is equally inside its normal channel. clear. In retrospect, the government allowing Lehman to fail on Sept. 15 was the trigger for the Euros fall, which was really the dollars rise. Fear and greed were replaced by fear and more fear, and the dollar became a safe haven. Now that the shock of the Lehman failure is fading, the Euro/dollar price is returning inside its normal channel. This is the basis, in part, of forecasts calling for the Euro to continue rising and for the price to meet 1.5000 around year-end and test the old July high of 1.6038 at some time during 2009.
Rethinking assumptions
This scenario might seem reaJanuary 2009 CURRENCY TRADER
14
FIGURE 2 SETTING UP A TEST If the Euro fails to surpass the previous intermediate high, it could test the low of 1.1815 from November 2005.
sonable, but its not. It assumes the Lehman Collapse Shock did not have a permanent effect on the primary trend. Once a trend is broken, we need to consider all factors anew. Notice the red support line connecting lows was broken along with the linear regression channel line. Okay, so lets assume the Lehman Shock started something fresh but its just not evident yet (Figure 2). Support and resistance get broken all the time and it doesnt necessarily mean a move is over, but we can feel
continued on p. 16 Source: data eSignal and Reuters Online; charts MetaStock
FIGURE 3 ALTERNATE SCENARIO An alternate scenario posits the recent EUR/USD high as an as-yet-untested breakout above the upper boundary of a new downtrend channel.
FIGURE 4 IMAGINARY TREND CHANNEL A hypothetical 45-degree downtrend channel projects the possibility of the EUR/USD pair challenging the October 2008 low, or even a run below 1.2000 by this time next year.
comfortable making such an assertion only if price now surpasses the previous intermediate high (upper horizontal gold line) at 1.4867, a level that occurred right after the Lehman Shock. If the Euro fails to match and surpass that level, it might test the previous low (bottom horizontal gold line) at 1.1815 from November 2005. In other words, there may be a new probable range forming. We can try to measure the new range by drawing a new trend channel (Figure 3). In this scenario, the recent EUR/USD high is a breakout, as yet untested, above the upper boundary of a new downtrend channel. Lets say the Euro falls by 50 percent of the original down move, or to about 1.3650. That would put it back at the channel top and, as we know, a 50-percent move is an important benchmark. Together with the failure to test and surpass the old intermediate high, it might mean the down-sloping channel is the right one. The problem with this channel is its a work in progress and its too steep. So lets invent a new more probable trendline channel at about a 45 degree angle, which is the slope of the up move from 2006 to the high in 2008 (Figure 4). Its fairly crazy to construct linear regression channels out of thin air, but lets add another assumption: high volatility and choppiness characterize the first phase of a crisis, and
January 2009 CURRENCY TRADER
16
FIGURE 5 BIG PICTURE FIBONACCI RETRACEMENT A 62-percent retracement of the move from 0.8229 in October 2000 to 1.6038 in July 2008 would put price at 1.1214.
as players become familiar with the new environment, increasingly stable prices will emerge. In fact, currencies sometimes have prolonged periods of sideways price action with little or no trend. This makes a trendlines slope less steep. If this scenario is true and useful, the EUR/USD pair could not only challenge the October 2008 low around 1.2329, but also make a run below 1.2000 by this time next year perhaps as far as 1.1000. In fact, 1.1214 is a 62-percent retracement of the bigpicture move from the low of 0.8229 in October 2000 to the high in July 2008 at 1.6038, as shown in Figure 5. Fibonacci numbers are an unfounded superstition, but enough traders like
continued on p. 18 Source: data eSignal and Reuters Online; charts MetaStock
them that we ignore them at our peril. If the imaginary trend channel is the correct scenario, why would the price development proceed that way? For once we have an easy answer. The European Central Bank (ECB) has been in denial, stating that inflation is still a worry and they need to see more data before they can cut rates further.
Now that the Fed has effectively cut to zero, the Bank of Japan has cut to nearly zero, and the Bank of England is widely expected to follow suit in early January, the ECB is the only major central bank that is out of line. The yield differential favors the Euro right now, with the overnight repo rate at 2.5 percent.
Evidently the ECB is not impressed by French wholesale inflation falling from 4.3 percent year-over-year in October to 1.9 percent in November, or other indicators of severe contraction. The German Kiel Institute says German GDP will fall 2.7 percent in 2009, to be followed by a pathetic 0.3 percent in 2010. One ECB policy mem-
18
ber said in December it would be logical for the bank to cut rates once it sees inflation expectations dropping under 2 percent. But the ECB is not entirely asleep at the switch. On Dec. 19 the ECB acted to remove some of the carry-trade charm of the Euro and halt a flood of incoming deposits from the dollar and the British pound by cutting the official deposit rate by 50 basis points to 1 percent below the key repo rate of 2.5 percent, and raising the lending rate by 50 basis points to 1 percent above the key rate. This created a wider rate spread corridor. It still leaves the Euro with a rate advantage for deposits, but a much smaller one, and it has the side effect of raising the cost of borrowing for European firms. Youd think thats the last thing a central bank would want to do in the current economic environment, although it has the bonus of raising bank profitability. The effect of the Dec. 19 announcement was immediate and huge the Euro swooned more than 500 points in a few hours. We must expect the same response if and when the ECB cuts the repo rate itself, possibly at its Jan. 8 policy meeting. The market expects the ECB to ultimately cut rates to 0.50 percent, a total of 200 basis points. If the ECB were to cut by 25-bp increments, it would have many months to go to get to 0-0.25 percent, like the Fed. In fact, it would take to Sept 2010. This seems improbable on the face of it. The ECB may be stubborn but it also values its reputation as a responsive institution, even if it values more highly its reputation for inflationfighting. So lets assume the ECB cuts by 50-bp increments, getting to 0.025 percent or 0.50 percent by June. The expectation of these cuts is Euro-negative.
CURRENCY TRADER January 2009
Finally, we all know by now that monetary policy alone cannot carry the weight of fixing a Great Depression II. Thats why the Fed followed the Bank of Japan and instituted a policy of quantitative easing, which in practice means buying just about any assets not on life-support from the banks and giving them cash. So far the banks are hoarding cash and refusing to lend it out, but Fed chairman Ben Bernanke has hope the banks will eventually find the confidence to start lending again. The ECB has already bought much in the way of unconventional collateralized assets, ballooning its balance sheet by roughly a third, but it does not have the ability to mandate stimulative fiscal policies. Together the EU countries will be spending 200 billion, but this is independent of the ECB. Europe lacks pan-EU fiscal institutions to boost every nation across the board. The collaboration of the U.S. Fed and the U.S. Treasury, together with an activist Executive, hold out more recovery hope than Europe can dream of probably a total of $2 or even $3 trillion. In fact, we might say the one thing the ECB lacks above all else is President-elect Barack Obama, although French President Nicolas Sarkozy has the same whatever it takes stance.
recovery and talk swirls around regarding when the Fed should start raising rates to offset all that inflationinducing new money supply. In short, an ECB being out of sync with the Fed has only a temporary Euro-favorable effect. Financial crisis has already morphed into an economic crisis, and the ECB lacks sufficient tools to deal with it. To be fair, its not certain the U.S. action will suffice to pull the country out of Great Depression II. But sitting on your hands and letting the generals fight the last war (inflation) sure wont do it, either. For information on the author see p. 6.
Advantage, dollar?
Last fall some clever analyst came up with the FIFO scenario the U.S. was the first in to financial institution crisis and recession, so it would also be the first out. Judging from the robustness of the U.S. government response, this is probably a good bet. June could arrive with the ECB still cutting rates and European governments squabbling over additional fiscal stimulus while the U.S. begins to show signs of
19
TRADING STRATEGIES
Inside-day setups
Fading the direction of inside days in the Euro/U.S. dollar pair shows promise, but trend direction makes a big difference in the results.
BY CURRENCY TRADER STAFF
were more often than not followed by bullish price action during the analysis period, but there was another, more interesting, tendency that appeared tradable in some situations: Inside days were often followed by inverted price action relative to the inside days close. Inside days that closed higher preceded short-term down moves, and vice versa. The following analysis shows how specific inside-day signals reflecting these tendencies performed over a 10-year period and offers insight on how to construct useful trade setups. The signals are applied to the EUR/USD pair because previous testing showed the basic pattern TABLE 1 FADING UP- AND DOWN-CLOSING INSIDE DAYS tendencies were more consistent and profitable in With the addition of a very simple filter, taking trades in the opposite direction of an inside this market. days close proved to be profitable on both the long and short sides of the market. Better
performance in different categories is highlighted. All trades no filter filter $11,647 $28,369 1.11 1.6 340 168 45.29% 54.76% $34 $169 $790 $823 -$591 -$623 1.34 1.32 9 7 9 5 $6,800 46.59% 21.69% 57 days $3,360 113.48% 11.42% 127 days Long trades no filter filter $13,799 $17,406 1.28 1.76 178 92 46.63% 54.35% $78 $189 $767 $809 -$525 -$548 1.46 1.47 5 4 9 5 $3,560 $1,840 Short trades no filter filter -$2,152 $10,963 0.96 1.45 162 76 43.83% 55.26% -$13 $144 $817 $840 -$661 -$715 1.24 1.17 7 5 7 5 $3,240 $1,520
he November and December 2008 issues of Currency Trader featured articles examining the short-term behavior of major currency pairs after inside days. The original analysis covered seven currency pairs: U.S. dollar/Canadian dollar (USD/CAD), Euro/U.S. dollar (EUR/USD), British pound/U.S. dollar (GBP/USD), U.S. dollar/Japanese yen (USD/JPY), U.S. dollar/Swiss franc (USD/CHF), Australian dollar/U.S. dollar (AUD/USD), and New Zealand dollar/U.S. dollar (NZD/USD). The results showed inside days overall, in all currency pairs
Net profit Profit factor Number of trades Winning percentage Avg. trade net profit Avg. winning trade Avg. losing trade Avg. win/avg. loss Longest winning streak Longest losing streak Total commission Return on initial capital % time in market (exposure) Longest flat period Max. drawdown % of initial capital Net profit as % of drawdown
Source: TradeStation
20
Strategy code
The rules for short trades are: Enter short at todays close if: 1. Todays low is above yesterdays low. 2. Todays high is below yesterdays high. 3. Todays close is above yesterdays close. Exit position at the close two days after entry. All trades were exited on the close two days after entry because initial pattern testing showed the early riskadjusted return was highest on that day: Further gains were probable with longer holding periods, but risk increased at a slightly faster pace. The setup was tested on daily data in the EUR/USD pair from Dec. 31, 1998 to Dec. 30, 2008. A nominal initial account value of $25,000 was used and $10 was assessed per trade for commission and slippage. Also, to see whether a trend filter might improve performance, the pattern was tested taking long trades only when the inside days close was above the close 40 days ago and taking short trades only when the close was below the close 40 days ago. This rule was not optimized in any way 40 days was simply a representative intermediate-term look-back period. 1. Execute long trade signals only if todays close is above the close 40 days ago (Close > Close[40]). 2. Execute short trade signals only if todays close is below the close 40 days ago (Close < Close[40]). The signals were tested with the filter because preliminary analysis indicated performance could be enhanced by accounting for trend direction an important factor, considering the uptrend that has dominated the EUR/USD pair for much of its existence. Table 1 compares the results of the signals with and without the filter. The no-filter results were profitable
continued on p. 22
The following TradeStation code is for the long-side version of the second entry signal. The code allows for customization of the strength or weakness of the days close (closethresh) and the look-back period for the trend filter (trendlength). inputs: trendlength(40), closethresh(.35); if High < High[1] and Low > Low[1] and (Close-Low)/(High-Low) <= closethresh and close > close[trendlength] then Buy this bar on close;
21
overall, thanks to the long trades; short trades actually lost money. Adding the filter made a dramatic difference: Despite cutting the number of trades in half, profitability more than doubled, the maximum drawdown was reduced by nearly twothirds, short trades became profitable, and the overall winning percentage increased by nearly 10 percentage points. The results were profitable, but only mildly so. But they support the earlier analysis that suggested there was potential in fading the direction of an inside days close in the EUR/USD pair as long as the trade is not fading the prevailing intermediate-term market direction. Now lets look at a parallel setup that defines up-closing and downclosing days a bit differently.
Source: TradeStation
TABLE 2 FADING STRONG- AND WEAK-CLOSING INSIDE DAYS The basic (unfiltered) version of this pattern performed much better than the first pattern and, again, the filter enhanced most performance measures. All trades no filter filter $22,905.00 $24,718.00 1.31 1.87 254 112 49.61% 59.82% $90.18 $762.33 -$571.48 1.33 7 10 $5,080.00 91.62% 15.68% 81 Days $220.70 $791.39 -$629.00 1.26 9 5 Long trades no filter filter $12,643.00 $13,546.00 1.34 2.03 130 56 47.69% 58.93% $97.25 $803.23 -$546.43 1.47 6 6 $241.89 $809.33 -$572.26 1.41 5 4 Short trades no filter filter $10,262.00 $11,172.00 1.29 1.74 124 56 51.61% 60.71% $82.76 $722.72 -$599.87 1.2 9 9 $2,480.00 $199.50 $773.97 -$688.32 1.12 8 3 $1,120.00
Net profit Profit factor Number of trades Winning percentage Avg. trade net profit Avg. winning trade Avg. losing trade Avg. win/avg. loss Longest winning streak Longest losing streak Total commission Return on initial capital % time in market (exposure) Longest flat period
Max. drawdown -$8,430.00 -$5,000.00 -$7,291.00 -$4,115.00 -$6,900.00 % of initial capital 33.72% 20.00% 29.16% 16.46% 27.60% Net profit as % of drawdown 271.71% 494.36% 173.41% 329.19% 148.72%
Source: TradeStation
22
As formulas, these rules are: 1. If High < High[1]; 2. Low > Low[1]; 3. (Close-Low)/(High-Low) <= .35; The rules are inverted for short trades. Figure 1 shows some recent signals. The setup was tested on daily data in the EUR/USD pair from Dec. 28, 1998 to Dec. 29, 2008, with the same account size and trade fees as the first test. Again, trades were exited after two days and the pattern was tested with and without the 40-day trend filter. Table 2 shows the results. The patterns unfiltered results are better than the first patterns, and in almost every aspect, the filter version again performed better, especially on a riskadjusted basis. The filter cut the number of trades in half (and halved commissions and market exposure) while net profit increased slightly. Also, the balance between the maximum number of consecutive winning trades and the maximum number of consecutive losing trades tilted toward the latter without the filter; the longest winning streak far outpaced the longest losing streak with the filter. In short, the patterns basic tendencies appear to be enhanced when long signals are traded when the market is up and short signals are traded when the market is down. Other methods of identifying the trend or immediate trade context have the potential to produce better results than the rudimentary rule used in this test. One interesting detail in both patterns tests is the filtered versions produced slightly larger average losing trades than the unfiltered versions. But the filtered versions much higher winning percentages and slightly larger winning trade values more than made up for this deficit. Table 3 compares a few key statistics for the filtered version of the second setup in different portions of the
CURRENCY TRADER January 2009 23
analysis window: Dec. 28, 1998 through Dec. 28, 2001; Dec. 29, 2001 through Dec. 29, 2004; Dec. 30, 2004 through Dec. 29, 2008. In terms of the number of trades and winning percentage, the results were fairly consistent from period to period, with the exception of the negative return for long
continued on p. 24
TABLE 3 PERIOD COMPARISON Results were relatively consistent from period to period. All trades Long trades Short trades 12/9812/0112/0412/9812/0112/042/9812/0112/04Period 12/01 12/04 12/08 12/01 12/04 12/08 112/01 12/04 12/08 Net profit $2,500.00 $12,270.00 $11,318.00 ($1,430.00) $9,350.00 $6,196.00 $3,930.00 $2,920.00 $5,122.00 Profit factor 1.2 3.16 2.36 0.58 3.73 2.08 1.45 2.3 2.99 Number of trades 37 30 41 7 23 25 30 7 16 Winning % 54.05% 60.00% 65.85% 42.86% 60.87% 64.00% 56.67% 57.14% 68.75% Longest winning streak 9 5 5 2 5 5 8 3 5 Longest losing streak 5 3 3 3 3 3 3 2 2
Source: TradeStation
trades from the December 1998 December 2001 window. Short trades outperformed long trades in terms of consistency and reliability, not net profit in two of the three periods. This is a good thing, as it indicates the setup is not merely hitching a ride on the back of the EUR/USDs upside bias during the majority of the analysis period. Also, the table makes clear the most recent four-year period (December 2004 December 2008) was not the best period for the setup, although it was very positive. For such a simple, robust setup, the results arent bad, and theres plenty of room to extract more value from the pattern. First, losses were not controlled all trades were exited after two days, win or lose. Testing indicated additional profit potential existed beyond this time horizon, but the probabilities of success decreased as time passed i.e., winning percentages and reward-risk ratios declined, although they remained favorable. However, taking partial profits at the two-day point (or at a price level corresponding to a high-probability profit target in the first days of the trade) and protecting the remainder of the position with a stop-loss could allow for additional profits with minimal risk. Table 4 shows the results of combining the two filtered versions of the patterns. The results are predictably something of a compromise.
24
Combining the signals produced a higher net profit, but most of the reward/risk measures were not as good as the second patterns statistics. All trades $51,427.50 1.69 280 55.71% $183.67 $805.38 -$598.48 1.35 10 9 $5,600.00 205.71% 12.11% 99 Days -$8,960.00 35.84% 573.97% Long trades $29,292.50 1.84 148 54.05% $197.92 $800.67 -$511.19 1.57 7 9 $2,960.00 Short trades $22,135.00 1.56 132 57.58% $167.69 $810.33 -$704.46 1.15 8 7 $2,640.00
Net profit Profit factor Number of trades Winning percentage Avg. trade net profit Avg. winning trade Avg. losing trade Avg. win/avg. loss Longest winning streak Longest losing streak Total commission Return on initial capital % time in market (exposure) Longest flat period Max. drawdown % of initial capital Net profit as % of drawdown
Source: TradeStation
Related reading
Inside days in the major currency pairs Currency Trader, November 2008. Analysis of inside days that occur after short-term price thrusts. Inside days: Part 2 Currency Trader, December 2008. This follow-up study digs deeper into inside days and focuses on the U.S. dollar/Canadian dollar and the Euro/U.S. dollar pairs. You can purchase and download past articles at http://store.activetradermag.com.
ADVANCED STRATEGIES
FIGURE 1 AUSTRALIAN DOLLAR During the AUDs pronounced uptrends, the trend oscillator not only turned negative but occasionally fell into oversold territory.
urrencies long have enjoyed a reputation for trending, and with good fundamental reasons. Not only do currencies reflect longterm national policies and tendencies that are slow, if not impossible, to change (really, should we expect Switzerland and Argentina to be confused at any point?), they reflect relative monetary policies that also tend to persist. Two cases in point: The U.S. dollar strengthened for almost five years in the first half of the 1980s after Paul Volcker instituted his policy of high interest rates. The greenback fell just as spectacularly, and for an even longer period of time, after May 2002 under the weight of a deliberate policy by the Federal Reserve to solve all economic problems with easy credit. We can throw darts at a world map, start narrating the history of the country hit, and arrive at pretty much the same conclusion: Currencies are capable of posting massive long-term trends. And as any position trader understands intuitively, almost any trading system or set of indicators works in a trend. Markets make indicators work, not vice-versa. Two questions arise, then. First, if this is the case then why do self-described trend-followers in currencies tend to have such mediocre performance (see Why currency traders should be humbler, May 2007 or Currencies and commitments, June 2008)? Second, which currencies are in fact trendiest? The first question will be dismissed curtly with this bit of doggerel: The trend is your friend, except for the bend in the end. Everyone can see the same trend, the trade gets crowded, and then it reverses in an execution vacuum capable of vaporizing in a matter of hours weeks of hard-won gains. Such is the life of a trend-follower. The second question will be addressed for a set of six major currencies: the Canadian and Australian dollars (CAD and AUD), the Japanese yen (JPY), the Swiss franc (CHF), the
January 2009 CURRENCY TRADER
26
FIGURE 2 CANADIAN DOLLAR Once the CADs bull run began, excess volatility dropped during nearly all periods of positive trend.
British pound (GBP), and the Euro (EUR). We will visit a set of minor currencies next month.
Trendiness
Trends are like Supreme Court Justice Potter Stewarts famous definition of obscenity (I know it when I see it). If a market is moving in a straight line with few retracements, we all can spot the trend. But defining it is difficult. Two accepted methods of defining when a market has serial correlation of returns, or a lowerthan-expected number of day-to-day sign changes in returns, are the Durbin-Watson and Wald-Wolfowitz tests, which indicate the markets are close to being random in distribution. This is visually counterintuitive, but just as hikers get lost when they stop trusting their compasses, traders can get lost when their lying
Almost any trading system or set of indicators works in a trend. Markets make indicators work, not vice-versa.
eyes get in the way of reality. Other venerated technical indicators of trendiness, such as Welles Wilders directional movement index (DMI) and its associated average directional movement index (ADX) do a good job confirming when you are in a trend, but they tend to be slow to capture excessive movements and abrupt but significant trend changes. Moreover, the commonly used 14-day DMI period is a parameterized time period. (Not that this does not work: I learned a good deal of technical analysis from a bombast who insisted on measuring every indicator against a simple 14-day moving average, and who took great glee when one complex tool after another failed to pass the test. There is a powerful lesson here.)
FIGURE 3 BRITISH POUND The GBPs excess volatility tends to surge much higher above 0.00 than it falls below it, and it also has the highest average level of any of the major currencies.
27
lator (red bars) only for trending days. The stronger the volatility-adjusted trend, the further away from zero the trend oscillator will be. In general, trend oscillator readings greater than 0.40 or less than -0.40, marked on the trend charts with grey lines, indicate a market is becoming overbought or oversold, respectively. The bottom charts in each figure depict the excess volatility (green bars) of each market for those days when the market is in a trending state. Excess volatility is the ratio of the implied volatility for three-month non-deliverable forwards to the high-low-close volatility. (In a small twist from past practice, we subtracted 1.00 from this ratio to depict it more intuitively as an oscillator around zero.) Excess volatility indicates the market is uncomfortable with the existing trend and is buying insurance in the form of options against its reversal. The more negative this measure is, the more the market is comfortable with the trend, and vice-versa.
where H, L, and C are high, low, and close, respectively. Once the MA is calculated, the trend is defined as the volatility-adjusted oscillator around this central tendency. In the construction of the index, the trends zero point occurs when price and AMA are equal:
Values of N in excess of 20 define a trending market, while those less than 11 define a sideways market and those from 11 through 20 define markets in transition. Figures 1-6 depict the daily high-low range for each of the six major currencies over all days, but show the trend oscil28
had a long and powerful uptrend between the beginning of 2003 and the end of 2007, but as was the case with the AUD, it had a large number of short-lived bona fide downturns. A trendfollowing trader could and by evidence did get knocked out of long positions numerous times during this trend. The CAD, like the AUD, broke sharply during the SeptemberOctober 2008 credit crunch. Its break was swift and severe enough to be done without any trend reversals. The excess volatility chart tells the real story, though. It remained quite high for the CAD during the tail end of its long bear market (extending through 2001), but once its bull run began, excess volatility dropped during nearly all periods of positive trend. Also like the AUD, its excess volatility fell sharply in the September-October 2008 sell-off. There must be something about speaking English that is related to currency trends, as the third trendiest currency amongst the majors is the British pound (Figure 3). This is a little surprising given the GBPs primary trade is not against the USD but rather against the Euro. And visually the price chart is far noisier than that of either the AUD or the CAD. But lets remember that point about hikers and their compasses and accept the data for what it is. The most interesting aspect of the British pound here is, like the Canadian dollar, how high the excess volatility was in 1999-2002 and then how it switched to a pattern where volatility spikes tended to mark tops in price. A data-mining trading system designer (which may be a redundancy) could back-fit a trading system to sell the GBP on these volatility spikes. In addition, the GBPs excess volatility is both highly asymmetric it tends to surge much higher over 0.00 than it falls below it and has the highest average level of any of the major currencies by far. Few of us would expect the Japanese yen (Figure 4) to rank very high on any measure of trendiness, and we are not disappointed in that regard. The JPY has remained in a fairly narrow trading range since 1999, but within that range we have seen several substantial trending moves tied to global financial crises and developments in the yen carry trade (see A closer look at the carry trade, July 2007). Moves in the trend oscillator outside of 0.40 tend to produce mean-reverting responses, and as befits a long-term trading range, these moves have been fairly symmetric. The excess volatility measure for the JPY was quite high in 2000-2001 as the Bank of Japan contemplated quantitative easing, which at the time was regarded as improbable. Once they went to the policy, excess volatility fell and remained in a narrow range until the credit crunch emerged in mid2007. When traders unwound yen carry trades in response, excess volatility fell and remained low except for a brief period in early September 2008. The market knew what it
CURRENCY TRADER January 2009
FIGURE 5 SWISS FRANC The Swiss francs reputation for long-running, pronounced trends changed after the advent of the Euro and the realignment of global currency trading into dollar bloc and Euro blocs.
wanted to do with the yen, which was to repurchase what it had borrowed, and proceeded to do so without further ado. The Swiss franc (see The Swiss francs commodity connection, October 2008) used to, along with the old Deutsche mark, have a reputation for long-running, pronounced trends. This changed after the advent of the Euro and the realignment of global currency trading into two broad currency blocs the dollar bloc and the Euro bloc (see The dollar index and firm exchange rates, December 2005). Even so, the CHF remained in a broad uptrend against the USD from mid-2001 until the September-October 2008 credit crunch (Figure 5). This is evidenced by a very large number of overbought spikes on the trend oscillator against but one oversold spike in mid-2005. Also, the CHF appeared to be very comfortable within its uptrend judging by its low excess volatility measure after the Swiss National Bank ceased cutting its LIBOR target rate in mid-2003. And like other currencies, its excess volatility broke after September 2008 as the USD strengthened. We finally come to the least trendy currency amongst the majors: the Euro. Even though it spent the first two-and-ahalf years of its history declining against the dollar and the next six years rallying before breaking severely during the September-October 2008 credit crunch it has had sufficient backing and filling to spend almost 40 percent of its life outside of a trending state. As the deepest and most liquid currency market in the world, the Euro tends to get very
continued on p. 30
29
FIGURE 6 EURO Because it tends to get very crowded at the end of a trend, the Euro can reverse suddenly (which is why trend-followers in currencies have such a poor track record). The excess volatility spikes since 2003 have signaled when these reversals are coming.
Related reading:
The rupee and emerging markets Currency Trader, December 2008. Analysis suggests Indias status as a global economic power is no accident. Nordic currency confusion Currency Trader, November 2008. Get a handle on the dynamics of the Northern European currencies. The Swiss francs commodity connection Currency Trader, October 2008. How can the Swiss currency be, of all things, a commodity currency? Franc-ly, my dear, I dont give a carry Currency Trader, September 2008. Investigating the Swiss franc carry trade, and what might change its dynamics. The short, awful life of the dollar carry trade Currency Trader, August 2008. The implications of the weak-dollar policy and the dollars roles as a funding currency. Currencies and commitments Currency Trader, June 2008. Find out what COT data conveys about forex price action.
SPOT CHECK
historic global economic crisis that drove money into the U.S. dollar as a safe haven. Figure 1 is a daily chart of the EUR/USD pair. After reaching a closing high of 1.5990 on April 22, the pair moved mostly sideways before the midJuly breakdown (it eclipsed 1.600 intraTABLE 1 EURO VS. THE DOLLAR: ANALYZING THE OTHER PLAYERS day on three occasions during this period). The pair had shed more than 22 perIn the depths of the 2008 financial panic, the Euro lost significant ground vs. cent by the time it closed at 1.2453 on the U.S. dollar. However, it mostly gained ground vs. other major currencies Nov. 20 a sell-off that might seem dra(except the yen): Through Dec. 17, it had double-digit gains vs. the other majors (ex-yen) not too far behind the dollars performance. matic were it not for the fact that the Euro lost nearly 30 percent vs. the Japanese Apr. 22-Nov. 20 Nov. 20-Dec. 17 Apr. 22-Dec. 17 yen (JPY) during the same period (Figure Euro/dollar -22.12% 15.80% -9.82% 2). Table 1 compares the percentage moves in the Euro and the U.S. dollar vs. Euro vs. yen -29.03% 8.42% -23.05% the other major currencies (yen, British Euro vs. pound 5.50% 9.75% 15.78% pound, Swiss franc, Canadian dollar, Australian dollar, and New Zealand dolEuro/Swiss -4.92% 1.34% -3.65% lar) during the EUR/USDs AprilEuro vs. Canada 0.17% 7.56% 7.74% November down move and subsequent Euro vs. Aussie 20.34% 0.56% 21.02% November-December rebound. Euro vs. kiwi 18.98% 1.96% 21.31% The overarching and intertwined themes during the fall financial panic Average: 1.84% 4.93% 6.53% were liquidation and repatriation: money o some observers, the Euros sharp decline vs. the dollar from July through November was a sea change the end of the Euros bull run despite the fact the move was triggered by a
Median: 2.84% 4.76% 11.76% Avg. (ex-yen) Med. (ex-yen) Dollar vs. yen Pound vs. dollar Dollar vs. Swiss Dollar vs. Canada Dollar vs. Aussie Dollar vs. kiwi Average: Median: Avg. (ex-yen) Med. (ex-yen) 8.01% 5.50% -9.04% 26.19% 22.08% 28.63% 35.34% 34.58% 22.96% 27.41% 29.36% 28.63% 4.23% 1.96% -6.92% -5.47% -12.49% -8.12% -15.19% -13.55% -10.29% -10.30% -10.97% -12.49% 12.44% 15.78% -15.33% 22.15% 6.84% 18.19% 25.51% 25.71% 13.84% 20.17% 19.68% 22.15%
Euro story has really been the Dollar story, or even the Dollar/yen story.
managers and investors got out of all kinds of assets and reverted money to home-country currencies. The Japanese yen a short-side favorite in forex carry trades because of Japans perennial lowinterest-rate environment was boosted dramatically. Not only did the yen skyrocket vs. the Euro, it was also the only major currency the buck lost ground
January 2009 CURRENCY TRADER
32
FIGURE 1 EURO/DOLLAR SEESAW The EUR/USD pair collapsed as money flooded into the safe-haven dollar during the fall financial panic, then rebounded sharply in late November.
against between April 20 and Nov. 20 (-9.04). Further inspection of Table 1 suggests the Euro story has really been the Dollar story, or even the Dollar/yen story: The Euro was battered much more vs. the yen than the greenback, and the EUR/USD pairs specific decline was driven by a singular economic event. In fact, the Euro gained ground against the Australian dollar, New Zealand dollar, and British pound, held its own against the Canadian dollar, and dropped moderately vs. the Swissie. These gains overall did not compare to the U.S. dollars huge surges against these currencies, but they do highlight the unique dynamic that was pitting the Euro vs. the dollar in a battle it could not win. The EUR/USD pair rebounded to the tune of nearly 16 percent from Nov. 20 to Dec. 17 on a closing basis, cutting the Euros total loss since Source: TradeStation April 22 to a little less than 10 percent. Also, during this NovemberFIGURE 2 THE YEN WILD CARD December rally, the Euro gained The EUR/JPY (top) and USD/JPY pairs both dropped as money managers ground vs. all the other major currenshort the yen had to cover their positions. The Euro lost nearly 30 percent vs. cies while the U.S. dollar gave back the yen between April and November, while the dollar which was itself much more (a median loss of 10.30 perhurtling to the upside vs. most currencies lost 9.04 percent. cent). Overall, the Euro and the U.S. dollar are not too far apart when assessing the entire April 22 to Dec. 17 period, especially when the yen wild card is removed from the picture. The Euro gained an average 12.44 percent vs. the five remaining currencies (median 15.78), while the dollar gained an average 19.68 percent (median 22.15 percent). In short, the performance underscores the EUR/USD move is about the dollar more than the Euro. But it raises the question of the importance of European Central Bank (ECB) interest-rate policy as the new year begins. The ECB has lagged other central banks in cutting interest rates in the face of the global economic slowdown. If it is forced to compensate, it could put a lid on the Euros strong rebound vs. the dollar, even if the dollars safeSource: TradeStation haven pop is over and done with.
CURRENCY TRADER January 2009 33
INTERNATIONAL MARKETS
CURRENCIES (vs. U.S. DOLLAR)
Current price vs. U.S. dollar
0.93593 1.39782 0.70465 0.58127 0.10538 0.69863 0.12921 0.01103 0.02923 0.02037 0.81999 0.03044 0.14677 0.12904 0.42653 1.46317 0.03431
Rank*
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Country
Currency
Swiss franc Euro Australian dollar New Zealand dollar South African rand Singapore dollar Swedish krona Japanese yen Thai baht Indian rupee Canadian dollar Taiwanese dollar Chinese yuan Hong Kong dollar Brazilian real British pound Russian ruble
1-month gain/loss
13.26% 10.39% 8.99% 7.74% 7.02% 6.40% 6.15% 4.35% 3.25% 2.16% 1.75% 1.57% 0.87% 0.02% -0.26% -3.51% -4.05%
3-month gain/loss
4.85% -0.80% -11.33% -13.62% -12.95% 0.24% -10.69% 16.94% -1.52% -3.73% -12.99% -2.44% 0.32% 0.23% -18.25% -17.78% -11.96%
6-month gain/loss
-4.60% -11.35% -26.30% -23.53% -17.12% -4.93% -22.56% 16.78% -3.08% -11.74% -16.41% -7.48% 0.49% 0.62% -31.62% -26.62% -19.55%
52-week high
1.0375 1.6038 0.9849 0.8214 0.1489 0.7434 0.1718 0.01148 0.03396 0.03974 1.0297 0.03335 0.14677 0.12904 0.6414 2.0397 0.04334
52-week low
0.813 1.2329 0.6005 0.519 0.0841 0.6512 0.1152 0.00891 0.0262 0.01843 0.768 0.02969 0.1367 0.1279 0.3751 1.4351 0.03267
Previous rank
15 5 3 16 1 10 13 12 9 2 4 6 7 8 14 17 11
ACCOUNT BALANCE
Rank
1 2 3 4 5 6 7 8 9 10 11 12
34
Country
Singapore Switzerland China Hong Kong Netherlands Taiwan Sweden Russia Germany Japan Canada Brazil
2007
41.395 65.534 379.162 22.796 55.891 25.402 25.903 72.543 175.371 195.904 25.603 10.253
Ratio*
27 15.8 11.7 11.2 7.4 6.8 6 5.9 5.4 4.5 1.8 0.8
2006
36.288 58.708 249.866 20.586 8.6 24.661 27.707 95.322 147.134 170.437 20.792 13.276
2008+
42.208 64.106 453.146 20.456 6.7 28.365 25.584 49.181 174.137 195.145 17.909 4.299
Rank
13 14 15 16 17 18 19 20
Country
Mexico France India UK Australia U.S. South Africa Spain
2007
-6.368 -39.363 -23.131 -96.687 -50.816 -784.341 -18.495 -138.916
Ratio*
-0.7 -1.6 -2.1 -3.5 -5.7 -5.7 -6.7 -9.8
2006
-2.425 -27.712 -9.503 -77.236 -41.49 -811.483 -16.608 -106.399
2008+
-10.588 -48.885 -32.301 -105.144 -52.988 -788.293 -19.237 -154.849
Totals in billions of U.S. dollars *Account balance in percent of GDP +Estimate Source: International Monetary Fund, World Economic Outlook Database, October 2008
January 2009 CURRENCY TRADER
1 Franc / Pound 2 Aussie $ / Pound 3 Franc / Canada $ 4 Franc / Yen 5 Aussie $ / Canada $ 6 Euro / Yen 7 Canada $ / Pound 8 Aussie $ / Yen 9 Franc / Euro 10 Aussie $ / Euro 11 Real / Canada $ 12 Canada $ / Yen 13 Real / Pound 14 Aussie $ / Franc 15 Real / Yen 16 Pound / Yen 17 Canada $ / Euro 18 Real / Aussie $ 19 Real / Euro 20 Pound / Euro
1 Mexico IPC 2 Brazil Bovespa 3 India BSE 30 4 Japan Nikkei 225 5 Hong Kong Hang Seng 6 Singapore Straits Times 7 UK FTSE 100 8 South Africa FTSE/JSE All Share 9 U.S. S&P 500 10 Germany Xetra Dax 11 Canada S&P/TSX composite 12 France CAC 40 13 Italy MIBTel 14 Switzerland Swiss Market 15 Australia All ordinaries Country U.S. Japan Eurozone UK Canada Switzerland Australia New Zealand Brazil Korea Taiwan India South Africa Rank 1 2 3 4 5 Country Germany UK U.S. Australia Japan
Last change 0.5 (Dec. 08) 0.2 (Dec. 08) 0.75 (Dec. 08) 1.00 (Dec. 08) 0.75 (Dec. 08) 0.5 (Dec. 08) 1.00 (Dec. 08) 1.50 (Dec. 08) 0.75 (Sept. 08) 1.00 (Dec. 08) 0.75 (Dec. 08) 1.00 (Jan. 09) 0.5 (Dec. 08) 3-month 8.26% 3.99% 7.62% 1.47% 1.67% 6-month 13.76% 4.60% 8.83% 2.79% 3.85%
35
ASIA AND SOUTH PACIFIC Australia Q3 Hong Kong Q3 India Q3 Japan Q3 Singapore Q3
Unemployment
Period AMERICAS Argentina Brazil Canada EUROPE France Germany UK Q3 Nov. Nov. Release date 12/22 12/19 12/5 Rate 7.8% 7.6% 6.3% 7.7% 7.1% 6.0% 1-year Change change -0.2% 0.1% 0.1% 0.1% 0.0% 0.2% -0.3% -0.6% 0.4% -0.5% -1.0% 0.7% Next release 2/25 1/22 1/9 3/5 1/7 1/21 Period Release date Rate 1-year Next Change change release
ASIA AND SOUTH PACIFIC Australia Nov. 12/6 Hong Kong Sept.-Nov 12/18 Japan Nov. 12/26 Singapore Q3 10/31
CPI
Period AMERICAS Argentina Brazil Canada EUROPE France Germany UK Nov. Nov. Nov. Nov. Nov. Nov. Release date 12/10 12/5 12/19 12/16 12/17 12/16 Change 0.4% 0.4% -0.3% -0.5% -0.5% -0.1% 1-year change 7.9% 6.4% 2.0% 1.6% 1.4% 4.1% Next release 1/13 1/9 1/23 1/14 1/15 1/20 AFRICA S. Africa Period Nov. Release date 12/17 Change 0.1% 1-year change 11.8% Next release 1/28
ASIA AND SOUTH PACIFIC Australia Q3 Hong Kong Nov. India Nov. Japan Nov. Singapore Nov.
PPI
Period AMERICAS Argentina Brazil Canada EUROPE France Germany UK Nov. Nov. Oct. Nov. Nov. Nov. Release date 12/10 12/8 11/28 12/22 12/19 12/8 Change -0.3% -0.2% -12.5% -1.9% -1.5% -0.7% 1-year change 9.8% 12.9% -0.2% 1.6% 5.3% 5.1% Next release 1/13 1/7 1/6 1/30 1/21 1/9 AFRICA S. Africa Period Nov. Release date 12/18 Change -1.3% 1-year change 12.6% Next release 1/29
ASIA AND SOUTH PACIFIC Australia Q3 Hong Kong Q3 India Nov. Japan Nov. Singapore Nov.
LEGEND: Change: Change from previous report release. NLT: No later than. Rate: Unemployment rate. As of Dec. 31
36 January 2009 CURRENCY TRADER
FOREX NEWS
he Fed cut its target federal funds rate on Dec. 16 to a range from 0 percent to 0.25 percent. Theyre acknowledging reality, says Joseph Trevisani, chief market analyst for New Jersey-based forex broker FX Solutions. Everybody knows the effective rate has been there for some time. The historic low came at a time when inflation was decreasing at record levels, bringing on new worries about the risk of deflation. November Consumer Price Index (CPI) data, also released on Dec. 16, showed a 1.7-percent drop. This drop was preceded by a 1.0-percent drop in October. The Bureau of Labor Statistics (BLS) attributes much of this drop to the
decline in energy prices. In November alone the BLSs energy index fell 17 percent, capping four months of decreases and doubling the October drop. The next-largest declines occurred in transportation, which fell 9.8 percent in November, and housing, which fell 0.1 percent. Nonetheless, the November CPI was 1.1 percent higher than the November 2007 reading. However, the 12-month change in July was at 5.6 percent, when crude oil prices were at their peak. Looking beyond the energy factor, Trevisani says the real type of deflation the Fed should be fearful of is the feedback loop. Prices fall and then people expect them to fall so they hold off purchas-
ing. This drops consumption, which drops employment, Trevisani says. Companies try to combat muted consumption by lowering prices and employing fewer workers. Unemployment in turn perpetuates less consumption, further fueling the
FIGURE 1 USD/JPY USD fell to a 13-year low vs. JPY the day after the Fed cut rates to its lowest level ever.
Managed money: Barclay Trading Groups currency trader rankings for November 2008
Top 10 currency traders managing more than $10 million as of Nov. 30, ranked by November 2008 return. Rank Trading advisor 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Auriel Currency 2X Fund Plimsoll Capital (Headwind) Goldman Sachs (Fund. Currency) John W. Henry & Co. (Int'l. FX) Sunrise Cap'l Partners (Currency Fund) Hathersage (Long Term Currency) Geo Economic Mgmt. System Ltd Spot Forex Mgmt. (Geneva) Capricorn Advisory Mgmt (fxST Aggres.) arsago Premium Currencies November return 5.76% 5.30% 4.65% 4.17% 2.57% 2.50% 2.29% 2.12% 2.05% 1.84% 2008 YTD return -15.56% 15.14% 20.67% 76.09% 16.35% 8.99% 16.37% 9.97% 5.88% 2.57% $ Under mgmt. (millions) 433.1 22.1 310.0 29.1 29.2 362.1 43.1 12.0 62.0 181.2
Source: eSignal
FIGURE 2 EUR/USD The Euro rose significantly against the dollar in the days surrounding the December Fed rate cut.
Top 10 currency traders managing less than $10 million and more than $1 million as of Nov. 30, ranked by November 2008 return. 1. Wallwood Consultants (Forex) 13.66% -42.66% 2. Spot Forex Mgmt. (Lausanne) 8.83% 46.81% 3. Spot Forex Mgmt. (Zurich) 4.36% 21.64% 4. Zone Cap'l FX Managed Account 4.18% 16.35% 5. Aspect Capital (Gl. Currency) 3.10% -4.26% 6. Swing Capital (FX) 2.03% 25.68% 7. Capricorn Advisory Mgmt (fxMT Growth) 1.62% 4.65% 8. Ketch Capital Management (Tack Fund) 1.54% 0.25% 9. Coe Capital Advisors (FX) 1.02% 5.53% 10. Quiddity (FX) 0.92% -6.99%
2.0 4.0 6.5 1.2 5.0 6.5 1.0 4.0 4.3 7.0
Source: eSignal
Source: BarclayHedge (http://www.barclayhedge.com). Based on estimates of the composite of all accounts or the fully funded subset method. Does not reflect the performance of any single account. PAST RESULTS ARE NOT NECESSARILY INDICATIVE OF FUTURE PERFORMANCE. 38
The information does NOT constitute trade signals. It is intended only to provide a brief synopsis of each markets liquidity, direction, and levels of momentum and volatility. See the legend for explanations of the different fields.
Market
OI
Volatility ratio/rank
Eurocurrency EC CME 153.0 124.9 -3.42% / 100% 9.27% / 62% 1.25% / 63% .16 / 43% Japanese yen JY CME 77.2 112.1 -4.58% / 100% 1.02% / 3% 9.60% / 37% .14 / 0% British pound BP CME 54.4 85.9 -6.44% / 100% -1.86% / 6% -17.45% / 78% .13 / 8% Swiss franc SF CME 27.8 30.6 -0.29% / 0% 12.54% / 71% 5.05% / 82% .37 / 77% Canadian dollar CD CME 26.1 60.1 -1.24% / 50% 3.87% / 67% -9.16% / 49% .08 / 0% Australian dollar AD CME 23.5 47.2 1.55% / 16% 10.03% / 83% -1.24% / 2% .16 / 58% Mexican peso MP CME 4.2 26.7 -4.30% / 92% -2.50% / 27% -12.17% / 27% .14 / 48% U.S. dollar index DX ICE 4.1 23.3 2.72% / 100% -4.32% / 43% 1.77% / 12% .09 / 0% New Zealand dollar NE CME 1.2 14.0 -0.80% / 50% 10.43% / 100% -7.24% / 16% .12 / 2% E-Mini eurocurrency ZE CME 2.4 1.8 -3.42% / 100% 9.27% / 62% 1.25% / 50% .16 / 43% Note: Average volume and open interest data includes both pit and side-by-side electronic contracts (where applicable). Price activity is based on pit-traded contracts.
LEGEND:
Volume: 30-day average daily volume, in thousands. OI: 30-day open interest, in thousands. 10-day move: The percentage price move from the close 10 days ago to todays close. 20-day move: The percentage price move from the close 20 days ago to todays close. 60-day move: The percentage price move from the close 60 days ago to todays close. The % rank fields for each time window (10-day moves, 20-day moves, etc.) show the percentile rank of the most recent move to a certain number of the
previous moves of the same size and in the same direction. For example, the % rank for 10-day move shows how the most recent 10-day move compares to the past twenty 10-day moves; for the 20-day move, the % rank field shows how the most recent 20-day move compares to the past sixty 20-day moves; for the 60-day move, the % rank field shows how the most recent 60-day move compares to the past onehundred-twenty 60-day moves. A reading of 100% means the current reading is larger than all the past readings, while a reading of 0% means the current reading is lower than the previous readings. Volatility ratio/% rank: The ratio is the short-term
volatility (10-day standard deviation of prices) divided by the long-term volatility (100-day standard deviation of prices). The % rank is the percentile rank of the volatility ratio over the past 60 days.
This information is for educational purposes only. Currency Trader provides this data in good faith, but assumes no responsibility for the use of this information. Currency Trader does not recommend buying or selling any market, nor does it solicit orders to buy or sell any market. There is a high level of risk in trading, especially for traders who use leverage. The reader assumes all responsibility for his or her actions in the market.
downward spiral. But for the moment, I dont see that happening, he says.
A down dollar
Both the Euro and the Japanese yen rallied for five straight days surrounding the Fed rate cut. From Dec. 11 through Dec. 17 the U.S. dollar (Figure 1) fell 5.9 percent against the yen, hitting a 13-year low on Dec. 17. On Dec. 19 the Bank of Japan lowered its interest rate to 0.1 percent, its lowest point since the country ended
its five-year zero-rate policy in 2006. December capped a year in which the U.S. dollar/Japanese yen (USD/JPY) fell 19 percent, the pairs largest annual drop in more than 20 years. The Euro (Figure 2) gained 10.7 percent against the dollar during the same time period, rallying 10 percent in December, the Euro/U.S. dollars (EUR/USD) largest monthly gain in 30 years. The December gain occurred after the Euro lost more than 20 percent against the dollar from July through November.
Outrageous predictions
n Dec. 17, London-based Saxo Bank released their annual outrageous claims for the year ahead, an attempt to predict rare but high-impact events that could occur in the year to come. Among the predictions for 2009 were that the EUR/USD pair could fall to 0.95 and then go as high as 1.30 as the European Central Bank
CURRENCY TRADER January 2009
attempts to deal with the economic troubles of its member states. Saxo also predicts 2009 could see the first Asian currencies pegged to the Chinese yuan, and that Asian economies will look toward China to find new trade partners and scale down their hitherto U.S.-centric agenda.
39
Tradingpatterns.com released version 7 of Automatic Pattern Search (APS), which has a redesigned search engine and results displays that include additional performance parameters. APS discovers trading systems based on price patterns that fulfill user-defined performance statistics and risk/reward parameters by searching historical data. Technical traders can use APS to develop and analyze stock, futures, and forex trading systems based on price patterns. APS generates code for implementing the price patterns it discovers in other popular programs, such as Metastock, Tradestation, Wealth-Lab, and TeleChart. ICAP, an interdealer broker, and Prism Valuation, a provider of mark-to-market and valuations services, final-
ized an agreement to allow Prism Valuation to use ICAP data as the main underlying source of OTC information for its services. The agreement covers a broad range of OTC data generated from ICAPs interdealer broker activities and will be used to assist in the valuation of complex structured products. Prism Valuation provides transparent valuations for complex OTC derivatives and structured products including interest rates, forex, inflation, equities, credit, commodities, and hybrids.
Note: New Products and Services is a forum for industry businesses to announce new products and upgrades. Listings are adapted from press releases and are not endorsements or recommendations from the Active Trader Magazine Group. E-mail press releases to editorial@currencytradermag.com. Publication is not guaranteed.
EVENTS
Event: The World Money Show Date: Feb. 4-7 Location: Orlando Date: March 17-19 Location: Hong Kong Date: May 11-14 Location: Las Vegas For more information: Go to http://www.moneyshow.com and click on Events Event: Live Trading Software and Trading Expos Event: International Traders Expo Date: Feb. 21-24 Location: New York, New York For more information: http://www.tradersexpo.com Event: The 15th Forbes Cruise for Investors Event: Securities Operations World 2009 Date: Feb. 24 Location: New York City For more information: http://www.fmwonline.com Event: 25th Annual Risk Management Conference Date: March 8-10 Location: The Ritz-Carlton, Laguna Niguel, Dana Point, Calif. For more information: http://www.cboermc.com Date: June 2-14 Location: Lisbon to Venice For more information: Go to http://www.moneyshow.com and click on Events Event: International Traders Expo Date: June 3-6 Location: Los Angeles For more information: http://www.tradersexpo.com Date: April 5-7 Location: Miami Date: June 7-9 Location: Houston For more information: http://livetradingexpo.com Event: Second Annual Conference on Institutional Options Trading Date: March 10 Location: New York City For more information: http://www.fmwonline.com
40
JANUARY/FEBRUARY
21
Brazil: Central Bank of Brazil monetary policy announcement Germany: December PPI
January 1 2 3 4 5 6
U.S.: December ISM non-manufacturing report and FOMC minutes Canada: November PPI U.S.: December ISM report
22
U.S.: December housing starts Brazil: December employment report Japan: Bank of Japan monetary policy announcement Hong Kong: December CPI Mexico: Jan. 15 CPI
23 24 25 26 27 28 29
7 8 9
Brazil: December PPI Germany: November employment report Mexico: Dec. 31 CPI and December PPI U.S.: December employment report Brazil: December CPI Canada: December employment report India: December PPI UK: December PPI LTD: January currency options (CME); U.S. Dollar Index futures (ICE)
U.S.: December leading indicators Australia: Q4 PPI Australia: Q4 CPI South Africa: December CPI U.S.: FOMC monetary policy statement and December durable goods Canada: December PPI Germany: December employment report South Africa: December PPI
10 11 12 13 14 15
U.S.: November trade balance U.S.: December retail sales France: December CPI U.S.: December PPI Australia: December employment report ECB: Monetary policy announcement Germany: December CPI Japan: December PPI
30
U.S.: Q4 GDP (adv.) and ECI France: December PPI India: December CPI Japan: December employment report and CPI
31 February 1 2 3 4 5
U.S.: December personal income Australia: Reserve Bank of Australia monetary policy announcement U.S.: January ISM report
January 2009 28 29 30 31 4 5 6 7 1 8 2 3 9 10
11 12 13 14 15 16 1 7 18 19 20 21 22 23 2 4 25 26 27 28 29 30 31 FEBRUARY 2009 1 8 2 9 3 4 5 6 7 10 11 12 13 14
16
17 18 19
Hong Kong: Oct.-Dec. employment report Mexico: December employment report
15 16 17 18 19 20 21 22 23 24 25 26 27 28
U.S.: January employment report Brazil: January CPI Canada: January employment report UK: January PPI LTD: February currency options (CME); February U.S. Dollar Index futures
The information on this page is subject to change. Currency Trader is not responsible for the accuracy of calendar dates beyond press time. January 2009 CURRENCY TRADER
20
41
KEY CONCEPTS
Average and median: The mean (or average) of a set of values is the sum of the values divided by the number of values in the set. If a set consists of 10 numbers, add them and divide by 10 to get the mean. A statistical weakness of the mean is that it can be distorted by exceptionally large or small values. For example, the mean of 1, 2, 3, 4, 5, 6, 7, and 200 is 28.5 (228/8). Take away 200, and the mean of the remaining seven numbers is 4, which is much more representative of the numbers in this set than 28.5. The median can help gauge how representative a mean really is. The median of a data set is its middle value (when the set has an odd number of elements) or the mean of the middle two elements (when the set has an even number of elements). The median is less susceptible than the mean to distortion from extreme, non-representative values. The median of 1, 2, 3, 4, 5, 6, 7, and 200 is 4.5 ((4+5)/2), which is much more in line with the majority of numbers in the set. Carry trades involve buying (or lending) a currency with a high interest rate and selling (or borrowing) a currency with a low interest rate. Traders looking to earn carry will buy a high-yielding currency while simultaneously selling a low-yielding currency.
The squared differences are used (instead of just the differences) because some differences are negative (for points below the line) and others are positive (for points above the line). Squaring all the differences creates all-positive values and allows you to calculate a formula for the straight line. The best-fit line is the line for which the sum of the squared differences between each price and the straight line are minimized. The formula for a straight line (y) is: y = a + b*t t = time a = the initial value of the line when t is equal to zero (sometimes called the intercept value i.e., the point at which the line intercepts the vertical y-axis) or the point at which a specific line begins b = the slope of the line, which is the rate at which the line rises or falls (e.g., 0.75 points per day). When fitting a straight line to N data points, the best-fit coefficients a and b can be solved for by: a = [2(2N+1)/N(N-1)] p(t) + [6/(N(N-1)] t*p(t)
t=1 t=1 N N
where,
Inside bar: A price bar with a lower high and higher low than the preceding bar. By definition, an inside bar represents a volatility contraction from the preceding bar. Linear regression (best-fit) line: A way to calculate a straight line that best fits a set of data (such as closing prices over a certain period) that is, a line that most accurately reflects the slope, or trend, of the data. A regression line is calculated using the least squares method, which refers to finding the minimum squared (x*x, or x 2) differences between price points and a straight line. For example, if two closing prices are 2 and 3 points away (the distance being calculated vertically) from a straight line, the squared differences between the points and the line are 4 and 9, respectively.
where, p(t) = the price at point t N = the number of prices we are using to calculate the coefficients. Volatility: The level of price movement in a market. Historical (statistical) volatility measures the price fluctuations (usually calculated as the standard deviation of closing prices) over a certain time period e.g., the past 20 days. Implied volatility is the current market estimate of future volatility as reflected in the level of option premiums. The higher the implied volatility, the higher the option premium.
42
pair
the most recent results of the strategy based on the setup described in Short-term momentum signals in the Euro (Currency Trader, May 2008), Euro momentum system, interrupted (Currency Trader, June 2008), and Euro momentum signal, tweaked (Currency Trader, July 2008). The system goes short when a short-term momentum calculation is strong relative to a bullish longer-term momentum calculation. In this case, the 10-day momentum indicator (ranging from -1.00 to +1.00) had to be greater than or equal to 0.80 while the 62day momentum was below -0.6. Note: These signals were not actually executed in the market, so the results are shown for the following parameters: a 0.0240-point profit target (a representative value found in testing) and a move above the high of the entry bar as the stop-loss. In both cases, the exits are executed on the close of the bar that the condition is triggered on.
Initial stop: A close above the high of the entry bar, estimated as 1.3081 (for the first two entries) and 1.3406 (for the third entry) in the Trade Summary table. Initial target: The first close 0.0240 or more below the entry price (estimated as 1.2754, 1.2864, and 1.2822 in the table).
Source: TradeStation
RESULT
Exit: 1.2689 (first trade and second trades); 1.3721 (third
trade).
Profit/loss: +0.0265 (first trade); +0.0375 (second trade); -0.0699 (third trade). Trade executed according to plan? Yes. Outcome: The first two trades were exited on Nov. 28 at a
better-than-estimated price, thanks to the relatively big down move and low close on that day. The third trade wiped out the accumulated profit and then some as the Euro began what turned out to be a very strong upside reaction that peaked in mid-December. While the approach of exiting on the close helped performance on the winning trades, it was disastrous in the case of the losing trade. Despite the two profitable trades, these results underscore a previously discussed weakness: The signals often come early and executing trades based on the opens and closes of the bars (rather than specific intrabar price levels) can put trades at a disadvantage. Past trades based on these signals from previous Trade Journals have used discretion to counter these problems. Further research should be conducted to see if these modifications can be applied systematically for future use. Also, like any signal of this type, the system is vulnerable to bad signals when a market transitions from uptrend to downtrend, or makes an outsized move in either direction.
Note: Initial trade targets are typically based on things such as the historical performance of a price pattern or a trading system signal. However, because individual trades are dictated by immediate circumstances, price targets are flexible and are often used as points at which to liquidate a portion of a trade to reduce exposure. As a result, initial (pre-trade) rewardrisk ratios are conjectural by nature.
TRADE SUMMARY
Date Contract Entry 1.2954 1.3064 1.3022 Initial stop 1.3081 1.3081 1.3406 Initial target 1.2754 1.2864 1.2822 IRR 1.18 11.76 0.52 Exit 1.2689 1.2689 1.3721 Date P/L LOP 0.0393 0.0503 LOL -0.0126 -0.0006 -0.0699 Trade length 3 days 2 days 2 days
Legend: IRR initial reward/risk ratio (initial target amount/initial stop amount); LOP largest open profit (maximum available profit during lifetime of trade); LOL largest open loss (maximum potential loss during life of trade).
CURRENCY TRADER January 2009 43