Today's action helps but does not yet confirm the bearish case. We will likely know if the high is in by 8:30am tomorrow after the GDP report. I seriously doubt if AMZN's positive earnings will be enough to pull the market back up tomorrow. And even if the GDP report is not as bad as expected, any pop is likely to fizzle, so caution is still advised. However, a weak close tomorrow on higher volume will go a long way to confirming that 1/28/09 was the swing and wave E high. For confirmation I will be looking for either a continuation of today's decline followed by a minimum two day pullback to setup a sell pivot or a break of the 1/20 to 1/23 lows to get 100% short.
Gold, the metal, continued its rally today and I am looking for the rally to continue as long as the US Dollar continues to correct the advance from the December low with a target around 930. However, based on the cycles I do not expect gold's rally to last much more than another week. I will be looking to exit my long gold position at the target or the day after a positive large range day.
Thursday, January 29, 2009
A Simple Method To Improve IBDs Market Calls
I was tooling around in free stuff at http://www.decisionpoint.com/ last night studying the thrust/trend model (TTM) that Carl Swenlin uses, and it occured to me that IBDs follow-through method is quite similar in concept to the thrust component of the TTM model.
What would the results be if we added a trend filter and a confirmation to IBDs market calls? I looked at a couple of different ideas, but the simplest seems to be the following:
Trend Filters
1. The macd of the SP500 or Wilshire 5000 (choose 1) must be greater than its signal line.
2. The macd of the market or stock being traded must be greater than its signal line.
Confirmation
1. The SP500 or Wilshire 5000 (choose 1) must continue higher the day(s) after a confirmed rally call and before a market in correction call.
2. The market or stock being traded must continue higher the day(s) after a confirmed rally call and before a market in correction call.
By adding the trend filter and confirmation to the Qs for 2008 using IBD market calls, the performance went from a CAGR of 29.30% to a CAGR of 45.85% - a substantial improvement. This brings IBDs performance in line with the other higher performing trend following systems.
Using the above, we would not be a buyer today because the macd is below the signal line for the SP500, Wilshire 5000 and for the Qs. Also, the markets do not appear to be continuing higher after yesterday's confirmed rally call. Should the trend filters turn positive and we have the markets moving higher before a market in correction call, then and only then, would we consider buying using this modified method.
I hope this helps those who have been whipsawed as many times over the years as I have using IBDs market calls.
What would the results be if we added a trend filter and a confirmation to IBDs market calls? I looked at a couple of different ideas, but the simplest seems to be the following:
Trend Filters
1. The macd of the SP500 or Wilshire 5000 (choose 1) must be greater than its signal line.
2. The macd of the market or stock being traded must be greater than its signal line.
Confirmation
1. The SP500 or Wilshire 5000 (choose 1) must continue higher the day(s) after a confirmed rally call and before a market in correction call.
2. The market or stock being traded must continue higher the day(s) after a confirmed rally call and before a market in correction call.
By adding the trend filter and confirmation to the Qs for 2008 using IBD market calls, the performance went from a CAGR of 29.30% to a CAGR of 45.85% - a substantial improvement. This brings IBDs performance in line with the other higher performing trend following systems.
Using the above, we would not be a buyer today because the macd is below the signal line for the SP500, Wilshire 5000 and for the Qs. Also, the markets do not appear to be continuing higher after yesterday's confirmed rally call. Should the trend filters turn positive and we have the markets moving higher before a market in correction call, then and only then, would we consider buying using this modified method.
I hope this helps those who have been whipsawed as many times over the years as I have using IBDs market calls.
Wednesday, January 28, 2009
Limited Upside
First, a couple of comments: 1) the advance GDP report comes out Friday, not tomorrow, and 2) I missed the news on the "Bad Bank" plan until this morning. I was out running my wife around on some errands. Currently, she can't drive much since she is suffering from severe neck pain due to a car accident and is waiting on approval for surgery. All of us will encounter stress in our lives and it can, most likely, will affect our trading. I have had my share of stress over the last few years, and I can tell you that having a sound trading plan helps me keep from making stupid mistakes, although I certainly make mistakes from time to time.
I do not trade off of news events, except when such news events are concurrent with and supportive of a valid system trading signal. The news from last night and today really does not clarify anything with respect to the trend, but it does fit the profile for an E wave, or perhaps a 2nd wave. At this point the market is behaving as expected. The Qs should not push much above 30.84 to 31.00. The move following that level should be a hard selloff that quickly retraces back to the 1/20 low or we will have to reconsider the current viewpoint.
IBD has scored this as follow-through day, but I would still remain cautious. So far, markets are following the script of the post-election year cycle as can be seen on http://www.seasonalcharts.com/. The typical pattern shows an initial upmove in January followed by a low in the 3rd week and another strong advance to finish the month. However, February is hard down. Whether or not, the Qs make a minor new high, a selloff is the expected outcome at this point. Last year the election year cycle inverted which gave excellent guidance for the hard down move in the fall, although it was a little early. Now, the cycle has inverted back to its typical pattern, and regardless or whether we see new lows are not, February is expected to be down, so I would not get too caught up in the current bullishness without a pullback and confirming breakout.
I keep wondering what creative new scheme our government will come up with next. I wish I could move my mortgage under an unrelated entity. The improvement in net worth would be nice. Really, this stuff is not that complex. The fact is that if it won't work for an individual or small business, then it probably won't work. This is all just smoke and mirrors. If our government used basic personal finance, we would not be in this mess.
I do not trade off of news events, except when such news events are concurrent with and supportive of a valid system trading signal. The news from last night and today really does not clarify anything with respect to the trend, but it does fit the profile for an E wave, or perhaps a 2nd wave. At this point the market is behaving as expected. The Qs should not push much above 30.84 to 31.00. The move following that level should be a hard selloff that quickly retraces back to the 1/20 low or we will have to reconsider the current viewpoint.
IBD has scored this as follow-through day, but I would still remain cautious. So far, markets are following the script of the post-election year cycle as can be seen on http://www.seasonalcharts.com/. The typical pattern shows an initial upmove in January followed by a low in the 3rd week and another strong advance to finish the month. However, February is hard down. Whether or not, the Qs make a minor new high, a selloff is the expected outcome at this point. Last year the election year cycle inverted which gave excellent guidance for the hard down move in the fall, although it was a little early. Now, the cycle has inverted back to its typical pattern, and regardless or whether we see new lows are not, February is expected to be down, so I would not get too caught up in the current bullishness without a pullback and confirming breakout.
I keep wondering what creative new scheme our government will come up with next. I wish I could move my mortgage under an unrelated entity. The improvement in net worth would be nice. Really, this stuff is not that complex. The fact is that if it won't work for an individual or small business, then it probably won't work. This is all just smoke and mirrors. If our government used basic personal finance, we would not be in this mess.
Qs Heading To 30.72
The Qs are breaking out to the upside this morning from a small b wave triangle that completed yesterday afternoon. The most likely scenario is that this is wave E of the larger wave(4) triangle. E waves tend to suck everyone end based on some perceived good news event. So far, while the earnings news has been bad with lowered outlooks, etc, the perception has been that they could have been worse. Also, the 4th Qtr GDP report comes out tomorrow. As long as it is not worse than expected this is likely to be viewed in a positive light. However, caution is advised as this is typical market behavior in wave E.
The possibility of this outcome is why I only went 50% short with my index allocation last week. Once, and if, the markets roll back over and reconfirm the downtrend will I increase my index short position.
The Qs could run all the way back up to the January 6 high, but as long as they remain below it by even one tick, then we should conclude that it is still wave E. The other evidence supporting this view is that the Dow, SP500 and the Transports all declined much further relative to the Qs and will likely not approach the January 6 high on this rally. So, I think the key here is the Qs. If they exceed the January 6 high, then something else is going on and the bias will shift to the long side.
Traders should also not become extremely bullish if there is a lack of downside follow-through. As I will discuss in future posts, such an outcome would actually be more bearish for the 2009 outlook than if we get a nearterm retest of the 2008 lows.
The possibility of this outcome is why I only went 50% short with my index allocation last week. Once, and if, the markets roll back over and reconfirm the downtrend will I increase my index short position.
The Qs could run all the way back up to the January 6 high, but as long as they remain below it by even one tick, then we should conclude that it is still wave E. The other evidence supporting this view is that the Dow, SP500 and the Transports all declined much further relative to the Qs and will likely not approach the January 6 high on this rally. So, I think the key here is the Qs. If they exceed the January 6 high, then something else is going on and the bias will shift to the long side.
Traders should also not become extremely bullish if there is a lack of downside follow-through. As I will discuss in future posts, such an outcome would actually be more bearish for the 2009 outlook than if we get a nearterm retest of the 2008 lows.
Tuesday, January 27, 2009
Market Becoming Overbought
A rare condition is developing with the 5 period RSI over 70 while the 14 period RSI is below 30. Given the lack of upward progress since last week, this is a strong indication that a sharp selloff is just around the corner. At this point, I expect that the broader market will continue to rally into Friday and possibly next Monday, although it need not do so. Only a sharp explosive rally that closes above the 50dema will alter the situation in my opinion.
Today was an NR7 day (narrowest range in the last 7 days) for the Qs and the SP-500. This may mean a sharp move in either direction is imminent. If it is to the upside, expect a fakeout.
RIMM moved up nicely today and CHKP made a 3 month high. There are signs that some sectors and stocks have bottomed even though the broader indexes may move to new lows. The next four weeks would be an ideal time to be filtering for strong relative strength stocks to buy after the market bottoms.
Today was an NR7 day (narrowest range in the last 7 days) for the Qs and the SP-500. This may mean a sharp move in either direction is imminent. If it is to the upside, expect a fakeout.
RIMM moved up nicely today and CHKP made a 3 month high. There are signs that some sectors and stocks have bottomed even though the broader indexes may move to new lows. The next four weeks would be an ideal time to be filtering for strong relative strength stocks to buy after the market bottoms.
Monday, January 26, 2009
Waiting For Resolution
There's little to add after today's action. We are still awaiting confirmation of wave c of D down in the triangle or wave 3 of (5) down. However, if it is going to go down into the January 31(30) turn date, then it must begin tomorrow. Otherwise, we can expect a continuation of the muted rally until this Friday and then a selloff into Feb 6. So far the Dow has moved perfectly with my projected Gann turn dates with a top on January 6, a bottom on January 23 and so a possible top January 30 or February 2 would fit the pattern well.
One thing about trend following systems that traders must understand is that conditions like we have experienced since November 21 can do significant damage to your account, and this may be perfectly within the expected performance of the system. It is not enough to have a trading system with a trend filter, and entry and exit signals. You must know the system's expectancy, payoff ratio, optimal heat(portfolio risk) and the expected maximum drawdown.
The psychology of trading has become a popular field of interest in the last few years, but I think that perhaps there is too much emphasis put on psychological solutions to perceived trading problems. As long as you have a plan, you are trading the plan, and you do not hesitate to take the trade that comes in accordance with your plan, then any problems you might be experiencing have nothing to do with your state of mind. Rather, if you perceive a problem, you need to do more analysis and testing of your system with the objective of understanding the performance of the system under all conditions and what the optimal heat level should be.
One system that I tested this weekend had an expectancy of 31.46%, a payoff ratio of 3.49 and produced a CAGR of 62% with 40% heat and 3 x margin. The maximum drawdown was 50%. The same system with 10% heat and no margin produced a CAGR of 18% with a maximum drawdown of 17%. Notice how the maximum drawdown seems to be correlated to the CAGR. This is what one should expect, which means that annualized returns in excess of 100% may eventually lead to a total account meltdown.
However, there is one other factor that significantly affects the expected maximum drawdown and that is the stability of monthly (or specific) period returns. The more stable the returns are the lower the expected maximum drawdown will be. One way to estimate the maximum drawdown in your account is to multiply the standard deviation of your last N months returns (N>=12) by 5.
Suppose your average (not CAGR) monthly return over the last 12 months is 10% with a standard deviation of 6%, then you can expect a 30% drawdown at some point. On the other hand, if the standard deviation was 2%, you could expect a drawdown of 10%. Clearly, stability of returns is the most important factor in determining your ability to sustain high annualized returns without a significant drawdown.
This will be one area that I will be diligently working on this year and I look forward to sharing the results with you.
One thing about trend following systems that traders must understand is that conditions like we have experienced since November 21 can do significant damage to your account, and this may be perfectly within the expected performance of the system. It is not enough to have a trading system with a trend filter, and entry and exit signals. You must know the system's expectancy, payoff ratio, optimal heat(portfolio risk) and the expected maximum drawdown.
The psychology of trading has become a popular field of interest in the last few years, but I think that perhaps there is too much emphasis put on psychological solutions to perceived trading problems. As long as you have a plan, you are trading the plan, and you do not hesitate to take the trade that comes in accordance with your plan, then any problems you might be experiencing have nothing to do with your state of mind. Rather, if you perceive a problem, you need to do more analysis and testing of your system with the objective of understanding the performance of the system under all conditions and what the optimal heat level should be.
One system that I tested this weekend had an expectancy of 31.46%, a payoff ratio of 3.49 and produced a CAGR of 62% with 40% heat and 3 x margin. The maximum drawdown was 50%. The same system with 10% heat and no margin produced a CAGR of 18% with a maximum drawdown of 17%. Notice how the maximum drawdown seems to be correlated to the CAGR. This is what one should expect, which means that annualized returns in excess of 100% may eventually lead to a total account meltdown.
However, there is one other factor that significantly affects the expected maximum drawdown and that is the stability of monthly (or specific) period returns. The more stable the returns are the lower the expected maximum drawdown will be. One way to estimate the maximum drawdown in your account is to multiply the standard deviation of your last N months returns (N>=12) by 5.
Suppose your average (not CAGR) monthly return over the last 12 months is 10% with a standard deviation of 6%, then you can expect a 30% drawdown at some point. On the other hand, if the standard deviation was 2%, you could expect a drawdown of 10%. Clearly, stability of returns is the most important factor in determining your ability to sustain high annualized returns without a significant drawdown.
This will be one area that I will be diligently working on this year and I look forward to sharing the results with you.
Thursday, January 22, 2009
Clarification - Finally
As we approach the end of intermediate wave (4) of primary wave 1 down, the picture is becoming increasingly clear. The action of the last 4 days has completed a small symmetrical triangle that is clearly evident on the intraday charts. For starters this means that the trend is clearly down, which agrees with our trend following systems. The reason is that a triangle always precedes the last wave of a movement. If the market goes up tomorrow, then the small triangle is part of an upward abc correction to be followed by another decline. If it goes down tomorrow, then the small triangle is wave b of D of the wave (4) triangle or wave b of C of an ending diagonal triangle. At this point, the most probable outcome is a bearish descending triangle is nearing completion with wave c of D down to come over the next few days. Once complete, we will see wave E up, which may last as long as 5 days. The current Gann turn dates that appear to be in play are January 31 and February 6. January 31+- should mark the end of wave D and February 6+- should mark the end of wave E.
If the November lows are taken out on the next decline, then we are in wave c of 3 of an ending diagonal triangle with waves 4 up and 5 down to go.
The least likely, yet possible scenario, if the market goes up tomorrow is that we will be going down in wave 3 of (5).
Interestingly, back in November and December I commented on the lack of equality in time between waves (2) and (4). The triangle pattern makes waves almost equal with wave (2) at 57 days and wave (4) around 70 days (by 2/6/09) for the Qs.
Traders should continue to look for the short side until there is resolution. While I continued to look for a breakout in January that never came, the trend models are short now and should stay that way for some time.
The above elliott wave analysis helps give a context for the market action and what we might expect, but the important point is to realize that the trend models brought us to a confirming conclusion without knowledge of the pattern. There are many different elliott wave patterns, and sometimes, the patterns are clear enough to trade, but knowledge of the pattern is not necessary for successful trading.
I am now and will remain 50% short during the completion of wave D down. I will exit short on approach to the November lows, and look for a new entry point on the next rally with an initial position of 50% short. As (and if) price rolls over after wave E up, I will be 100% short.
The triangle pattern projects a low in late February or early March with the Qs retesting the 2002 lows at or below 19.72. Wave E should retrace back up to today's highs or slightly higher offering traders with a chance for another 30% gain in wave 5 down.
If the November lows are taken out on the next decline, then we are in wave c of 3 of an ending diagonal triangle with waves 4 up and 5 down to go.
The least likely, yet possible scenario, if the market goes up tomorrow is that we will be going down in wave 3 of (5).
Interestingly, back in November and December I commented on the lack of equality in time between waves (2) and (4). The triangle pattern makes waves almost equal with wave (2) at 57 days and wave (4) around 70 days (by 2/6/09) for the Qs.
Traders should continue to look for the short side until there is resolution. While I continued to look for a breakout in January that never came, the trend models are short now and should stay that way for some time.
The above elliott wave analysis helps give a context for the market action and what we might expect, but the important point is to realize that the trend models brought us to a confirming conclusion without knowledge of the pattern. There are many different elliott wave patterns, and sometimes, the patterns are clear enough to trade, but knowledge of the pattern is not necessary for successful trading.
I am now and will remain 50% short during the completion of wave D down. I will exit short on approach to the November lows, and look for a new entry point on the next rally with an initial position of 50% short. As (and if) price rolls over after wave E up, I will be 100% short.
The triangle pattern projects a low in late February or early March with the Qs retesting the 2002 lows at or below 19.72. Wave E should retrace back up to today's highs or slightly higher offering traders with a chance for another 30% gain in wave 5 down.
Subscribe to:
Posts (Atom)