The lesson? Sometimes market behaviors change.
Thursday, December 30, 2010
Nasdaq Performance on the Last Day of the Year
A subscriber inquiry brought this changing tendency to my attention and led me to investigate further. Below we see the last day of the year performance for the Nasdaq Composite.
The lesson? Sometimes market behaviors change.
The lesson? Sometimes market behaviors change.
When SPY Closes Near the Bottom of its Range but Still Positive
Yesterday's late selloff was something that many chart readers might view as ugly on a chart. My research has shown quite the opposite. When SPY has closed near the bottom end of its range but still positive on the day that has generally been a good thing. Below is a simple study from last night's Subscriber Letter that exemplifies this.
Even though the number of instances is near the low end of what I prefer the results are strongly suggestive of an upside edge. The profit factor and winning % are especially compelling.
Below is an equity curve using a 3-day exit strategy.
Equity curves don’t get much straighter or more attractive than this. In one of my next few posts I’ll be discussing some of the things I look for in a study that make it compelling. This one has numerous compelling aspects and will act as a nice example.
Even though the number of instances is near the low end of what I prefer the results are strongly suggestive of an upside edge. The profit factor and winning % are especially compelling.
Below is an equity curve using a 3-day exit strategy.
Equity curves don’t get much straighter or more attractive than this. In one of my next few posts I’ll be discussing some of the things I look for in a study that make it compelling. This one has numerous compelling aspects and will act as a nice example.
Thursday, December 23, 2010
Quantifiable Edges Big Time Swing System Overview Page Updated
I’ve updated the Quantifiable Edges Big Time Swing System overview page with results through December 21st. There is not a trade currently open and it’s unlikely we’d see a trade open and close before the end of the year so I figured I might as well do it now. I don’t update results that often since the system only trades about once per month on average. While 2010 was a subpar year, I am pleased to report that on a total of 12 trades it did post a little over a 4% gain.
It’s important not to overreact to a small sample of trades and any single year with this system is a small sample. So I’m not terribly concerned that the performance was subpar. 2010 was marked with moves that were more persistent than usual. Examples would be the March-April rally, the September-October rally and the recent December rally – all of which plugged forward without the sort of oscillations that are typically seen. For the Big Time Swing, which often looks to play oscillations, this meant some extended sidelined periods. There has only been 1 trade in the 4th quarter.
Profits were also cut in half thanks to a few positions that signaled an exit for the next morning. Exits can be taken at the close or the next day’s open. Historical analysis has shown an edge in holding certain trades overnight after the exit is triggered. Doing so in 2010 would have cost about 4%, so this did cause me some frustration. Still, I’m not inclined to change my approach due to a small number of unfortunate overnight moves. Of course since it is an open system traders have the option of tweaking it any way they want.
For those looking for a system that they can use as a base to build their own system from, the Big Time Swing is an attractive option. It is all open-coded and comes complete with a substantial amount of background historical research. And since it is only in the market about ¼ of the time, it can easily be combined with other systems to provide greater opportunities. Once you’re ready to try and improve the system yourself you can also refer to the system manual or the August 2010 purchaser-only webinar – both of which discuss numerous ideas for customization.
And if system development isn’t your thing, the Big Time Swing System provides easy to follow mechanical rules that you can follow. The standard parameters have performed quite well. There are only about 12 trades per year averaging 7 trading days per trade. All entries and exits are either at the open or the close. And to be sure you have everything set up properly traders may follow the private-purchasers only blog that shows all SPY signals and possible entry/exit levels. This service is free for 12 months from the date of purchase.
For more information and to see the updated overview sheet, click here.
If you’d like additional information about the system, or have questions, you may email BigTimeSwing @ Quantifiable Edges.com (no spaces).
It’s important not to overreact to a small sample of trades and any single year with this system is a small sample. So I’m not terribly concerned that the performance was subpar. 2010 was marked with moves that were more persistent than usual. Examples would be the March-April rally, the September-October rally and the recent December rally – all of which plugged forward without the sort of oscillations that are typically seen. For the Big Time Swing, which often looks to play oscillations, this meant some extended sidelined periods. There has only been 1 trade in the 4th quarter.
Profits were also cut in half thanks to a few positions that signaled an exit for the next morning. Exits can be taken at the close or the next day’s open. Historical analysis has shown an edge in holding certain trades overnight after the exit is triggered. Doing so in 2010 would have cost about 4%, so this did cause me some frustration. Still, I’m not inclined to change my approach due to a small number of unfortunate overnight moves. Of course since it is an open system traders have the option of tweaking it any way they want.
For those looking for a system that they can use as a base to build their own system from, the Big Time Swing is an attractive option. It is all open-coded and comes complete with a substantial amount of background historical research. And since it is only in the market about ¼ of the time, it can easily be combined with other systems to provide greater opportunities. Once you’re ready to try and improve the system yourself you can also refer to the system manual or the August 2010 purchaser-only webinar – both of which discuss numerous ideas for customization.
And if system development isn’t your thing, the Big Time Swing System provides easy to follow mechanical rules that you can follow. The standard parameters have performed quite well. There are only about 12 trades per year averaging 7 trading days per trade. All entries and exits are either at the open or the close. And to be sure you have everything set up properly traders may follow the private-purchasers only blog that shows all SPY signals and possible entry/exit levels. This service is free for 12 months from the date of purchase.
For more information and to see the updated overview sheet, click here.
If you’d like additional information about the system, or have questions, you may email BigTimeSwing @ Quantifiable Edges.com (no spaces).
Wednesday, December 22, 2010
'Twas 3 Nights Before Christmas (Updated)
Tuesday's close brought us to the next extremely strong seasonal period. The last 2 years I have shown the "Twas 3 Nights Before Christmas" study. I've have updated it again below.
Results continue to look strong. A close above the entry at some point in the next 5 days has been a near certainty since 1987.
Results continue to look strong. A close above the entry at some point in the next 5 days has been a near certainty since 1987.
Friday, December 17, 2010
How many instances are needed when considering study results?
This post is the 2nd part of a series I started a few weeks ago that will discuss using quantifiable edges to your advantage. Today I'll discuss a common question I get about the studies. How many instances are needed for valid and usable results? It will lead into "What makes a study compelling?" in the next post.
Many of the posts I put on the blog are what I refer to as studies. In this previous post I showed the layout of the studies. A study is simply test results of an idea. Most of the time the idea is based in technical analysis. It looks to answer the question, “How has the market performed in the past after…”
Some studies are fairly general. For instance, I might look at how the market performs after it has traded down 3 days in a row. Others are more specific with added filters. Perhaps I notice that not only is the SPX down 3 days in a row, but it also is trading at a 10-day low, and is above the 200ma and volume has increased each of the last 3 days.
Both studies could tell me something about the market in relation to its current condition (assuming I’m describing current conditions, which is typically my approach). If I am able to describe conditions that more closely match the current market then I have a better shot at seeing behavior over the next several days match up with the study results. Of course there is a trade-off between general and specific, and that is the number of instances.
A general test may have hundreds or thousands of instances which it can refer to in order to generate expectations. A very specific test may have an extremely low number of instances. If the number of instances is too low then the results may have little or no meaning. For instance if my parameters are run and I find that the market had only set up in a similar manner 1 other time over my test period, is it reasonable to assume that the market will act the same way this time? Most people would correctly assume “no”. What if there were 2 instances and they both had similar reactions in the past. Could I assume this suggests a directional edge? 3 instances? 4? 10? 30? 50? More? How many instances is “enough” to have some level of confidence that your results are actually suggesting an edge and they are not the result of luck?
Before answering let me address 1 common misconception people have about statistical testing. That misconception is that you need 30 instances in order to demonstrate statistical significance. This idea originates in the fact that a sample size of 30 is needed in order to calculate a Z-score or run a chi-square test. The reason that 30 instances are necessary is that Z-scores assume a normal probability distribution. Without 30 instances it is not possible to resolve the shape of the normal probability distribution clearly enough to make certain statistical measures valid. One thing traders should be aware of is that the stock market does not have a normal distribution anyway. It has “fat tails”. In other words, there are more outliers present in stock market movements than one would expect under a normally distributed curve. So relying on standard statistical measures and assuming a normal distribution could expose a trader to more risk than his results would imply.
Still, these tests are helpful in determining whether your results were likely due to a real edge or whether there is a high risk that luck played a big part. But what if you don’t have 30 instances? In that case you could use a t-table statistic.
To better understand statistical significance and see how to run some of these tests I’ll refer you to the below post from a couple of years back:
http://quantifiableedges.blogspot.com/2008/05/significance.html
Note that this post also contains a t-table. One interesting thing we can see when looking at a t-table is the minimum number of instances you would need to have different confidence levels that your edge is actually an edge and not due to luck. For instance, if all instances were followed by a market rise, you would want at least 6 instances in order to be 95% confident that there was an actual edge. A 99.9% confidence would be reached if you had 11 instances that all resulted in a rise over the next X days.
So if you look back at the study I showed Wednesday, SPY only set up in that pattern 12 times in the past, but every time it was trading higher 5 days later. This means statistically there is about a 99.9% chance that the positive results were due to more than luck. That there has in fact been a real edge in that pattern in the past. Does this mean there is a 100% chance it will be higher 5 days after the setup? No! Not even close. A high degree of confidence means there is likely some kind of an edge. It doesn’t mean the past winning % or net expectations are likely to persist indefinitely.
So how many instances do I require before I’m willing to accept a study as part of my analysis and place it on my active list? It varies depending on things like the strength of previous reactions and other stats I’ll get into in my next post, but I’ll generally use a t-table to help me decide. Will I incorporate a study with only 10 or 11 instances? Yes, but it will have to have strong win/loss stats and a high win %. Personally, I tend to favor studies that have somewhere between 20-70 instances. Too low and they are less reliable. Too high and the setup is often too broad to have much meaning.
I’ve spent far more space discussing this than I wanted, but it is an issue that has come up time and again with readers, so I wanted to be somewhat thorough.
In fact, of the list of things I look at in a study to help me decide whether it is compelling or not, the number of instances (assuming it isn’t minuscule) is near the bottom .
I intend to accelerate this series of posts over the next couple of weeks and I’m sorry it’s taken so long to get rolling. In the next post I will discuss a list of other things I examine when determining whether I find a study compelling.
Many of the posts I put on the blog are what I refer to as studies. In this previous post I showed the layout of the studies. A study is simply test results of an idea. Most of the time the idea is based in technical analysis. It looks to answer the question, “How has the market performed in the past after…”
Some studies are fairly general. For instance, I might look at how the market performs after it has traded down 3 days in a row. Others are more specific with added filters. Perhaps I notice that not only is the SPX down 3 days in a row, but it also is trading at a 10-day low, and is above the 200ma and volume has increased each of the last 3 days.
Both studies could tell me something about the market in relation to its current condition (assuming I’m describing current conditions, which is typically my approach). If I am able to describe conditions that more closely match the current market then I have a better shot at seeing behavior over the next several days match up with the study results. Of course there is a trade-off between general and specific, and that is the number of instances.
A general test may have hundreds or thousands of instances which it can refer to in order to generate expectations. A very specific test may have an extremely low number of instances. If the number of instances is too low then the results may have little or no meaning. For instance if my parameters are run and I find that the market had only set up in a similar manner 1 other time over my test period, is it reasonable to assume that the market will act the same way this time? Most people would correctly assume “no”. What if there were 2 instances and they both had similar reactions in the past. Could I assume this suggests a directional edge? 3 instances? 4? 10? 30? 50? More? How many instances is “enough” to have some level of confidence that your results are actually suggesting an edge and they are not the result of luck?
Before answering let me address 1 common misconception people have about statistical testing. That misconception is that you need 30 instances in order to demonstrate statistical significance. This idea originates in the fact that a sample size of 30 is needed in order to calculate a Z-score or run a chi-square test. The reason that 30 instances are necessary is that Z-scores assume a normal probability distribution. Without 30 instances it is not possible to resolve the shape of the normal probability distribution clearly enough to make certain statistical measures valid. One thing traders should be aware of is that the stock market does not have a normal distribution anyway. It has “fat tails”. In other words, there are more outliers present in stock market movements than one would expect under a normally distributed curve. So relying on standard statistical measures and assuming a normal distribution could expose a trader to more risk than his results would imply.
Still, these tests are helpful in determining whether your results were likely due to a real edge or whether there is a high risk that luck played a big part. But what if you don’t have 30 instances? In that case you could use a t-table statistic.
To better understand statistical significance and see how to run some of these tests I’ll refer you to the below post from a couple of years back:
http://quantifiableedges.blogspot.com/2008/05/significance.html
Note that this post also contains a t-table. One interesting thing we can see when looking at a t-table is the minimum number of instances you would need to have different confidence levels that your edge is actually an edge and not due to luck. For instance, if all instances were followed by a market rise, you would want at least 6 instances in order to be 95% confident that there was an actual edge. A 99.9% confidence would be reached if you had 11 instances that all resulted in a rise over the next X days.
So if you look back at the study I showed Wednesday, SPY only set up in that pattern 12 times in the past, but every time it was trading higher 5 days later. This means statistically there is about a 99.9% chance that the positive results were due to more than luck. That there has in fact been a real edge in that pattern in the past. Does this mean there is a 100% chance it will be higher 5 days after the setup? No! Not even close. A high degree of confidence means there is likely some kind of an edge. It doesn’t mean the past winning % or net expectations are likely to persist indefinitely.
So how many instances do I require before I’m willing to accept a study as part of my analysis and place it on my active list? It varies depending on things like the strength of previous reactions and other stats I’ll get into in my next post, but I’ll generally use a t-table to help me decide. Will I incorporate a study with only 10 or 11 instances? Yes, but it will have to have strong win/loss stats and a high win %. Personally, I tend to favor studies that have somewhere between 20-70 instances. Too low and they are less reliable. Too high and the setup is often too broad to have much meaning.
I’ve spent far more space discussing this than I wanted, but it is an issue that has come up time and again with readers, so I wanted to be somewhat thorough.
In fact, of the list of things I look at in a study to help me decide whether it is compelling or not, the number of instances (assuming it isn’t minuscule) is near the bottom .
I intend to accelerate this series of posts over the next couple of weeks and I’m sorry it’s taken so long to get rolling. In the next post I will discuss a list of other things I examine when determining whether I find a study compelling.
Wednesday, December 15, 2010
A Rare SPY Pattern That Has Always Been Followed by Short Term Gains
The pattern of the last 2 days is quite interesting. Both days we saw a gap higher, a move up above the previous day’s high, and then a reversal that led the SPY to close below its open but still in positive territory. I looked at this 2-day setup in the subscriber letter in March using a long-term trend filter. I have updated the study below.
Only 12 instances but the results are overwhelmingly positive. In last night's Subscriber Letter I shared some additional details, including all the dates. There are actually a very large number of studies I am currently monitoring. They are somewhat mixed. This particular study makes a compelling arguement for a short-term bullish outlook. If you'd like to trial the Quantifiable Edges Subscriber Letter a free trial is offered here. If you have already trialed it but not in the last 6 months, you may request another trial via email to support at QuantifiableEdges dot com.
Only 12 instances but the results are overwhelmingly positive. In last night's Subscriber Letter I shared some additional details, including all the dates. There are actually a very large number of studies I am currently monitoring. They are somewhat mixed. This particular study makes a compelling arguement for a short-term bullish outlook. If you'd like to trial the Quantifiable Edges Subscriber Letter a free trial is offered here. If you have already trialed it but not in the last 6 months, you may request another trial via email to support at QuantifiableEdges dot com.
Monday, December 13, 2010
I'll Be Speaking at the Traders Expo in New York in February
The Traders Expo will be held at the Marriot Marquis Hotel in from February 20 - 23, 2011. I've decided to make the trip.
I'll be speaking on the 21st from 1:30 - 2:30pm. I'll be discussing some of of my favorite research and trading ideas. I hope to have the opportunity to meet severall blog readers and subscribers at the event.
I'll send out another reminder as we get closer. Registration is free and you may sign up using the link below:
https://secure.moneyshow.com/msc/nyot/registration.asp?sid=nyot11&scode=020867
I'll be speaking on the 21st from 1:30 - 2:30pm. I'll be discussing some of of my favorite research and trading ideas. I hope to have the opportunity to meet severall blog readers and subscribers at the event.
I'll send out another reminder as we get closer. Registration is free and you may sign up using the link below:
https://secure.moneyshow.com/msc/nyot/registration.asp?sid=nyot11&scode=020867
Sunday, December 12, 2010
The Most Wonderful Tiiiime of the Yearrrrrr!
Over several time horizons op-ex week in December has been the most bullish week of the year for the SPX. The positive seasonality actually has persisted for up to 3 weeks. I demonstrated this last year in the 12/14/09 blog. I’ve updated that study below to include 2009 stats.
Last year saw the market move higher on Monday and then pull back the rest of the week before rallying into year-end.
I'm generally seeing a mix of bullish and bearish studies right now. Friday's blog is an example of an active bearish study. This one certainly favors the bulls.
Last year saw the market move higher on Monday and then pull back the rest of the week before rallying into year-end.
I'm generally seeing a mix of bullish and bearish studies right now. Friday's blog is an example of an active bearish study. This one certainly favors the bulls.
Friday, December 10, 2010
SPY Consecutive 50-day Highs On Lower Volume
Declining volume at new highs can often lead to short-term difficulties. Below is a study related to SPY and SPY volume that I've shown a few times in the Subscriber Letter. It popped up in the Quantifinder again on Thursday.
This appears to suggest a mild downside edge. The high probability of some kind of decline despite the fact that it always occurs in an intermediate-term uptrend makes the study compelling enough to me to take under consideration.
This appears to suggest a mild downside edge. The high probability of some kind of decline despite the fact that it always occurs in an intermediate-term uptrend makes the study compelling enough to me to take under consideration.
Tuesday, December 7, 2010
Large Gap to New Highs Not the Edge They Once Were?
I've shown in the past using numerous studies that a large gap to a new high has a tendency to pull back during the day. Below is a study that represents some of what I was looking at this morning.
The stats here appear quite bearish. But below is the equity curve.
It appears over the last few years this setup has failed to deliver consistent downside movement. I looked at this a number of ways this morning and most of the equity curves looked like this. So be careful getting overcondfident trying to short this gap.
The stats here appear quite bearish. But below is the equity curve.
It appears over the last few years this setup has failed to deliver consistent downside movement. I looked at this a number of ways this morning and most of the equity curves looked like this. So be careful getting overcondfident trying to short this gap.
Monday, December 6, 2010
POMO Stimulus Indicator At New High and Still Climbing
Last week on the blog I showed an indicator that measured the amount of POMO stimulus the Fed has injected into the system over a 1-month (20 day) timeframe. As a review POMO stands for Permanent Open Market Operations and it is how the Fed goes into the open market to buy (or sell) treasury securities. The net effect of this buying is an influx of cash into the system. It appears a portion of that cash makes its way through the banking system and into the stock market. It also appears that the net effect of all this Fed buying is a positive influence on the stock market.
Today I have updated the chart from last week. The top panel shows the S&P 500. The indicator on the bottom is the total POMO buying in dollars that the Fed has done. I've zoomed in to just show the last year and a half.
As you can see the POMO buying over the last month has now far exceeded any 20-day period in 2009 (or ever). According to the Fed's website Mon-Thurs of this week are also scheduled for POMO activity. And a new schedule is due out on Friday so there is a chance we'll continue to see strong Fed buying in the weeks ahead. Evidence suggests to me that this should have a bullish influence on the market.
Wednesday, December 1, 2010
Large Gaps Up on the 1st Day of the Month
Last night in his video "Gap Guy" Scott Andrews took some interesting looks at gaps that occured on the 1st day of the month. Of course since he posted it early he couldn't have known we were about to get such a large gap up. Inspired by Scott I took a look at other 1% gaps that occured on the 1st day of the month. I've listed all instances below.
Subscribe to:
Posts (Atom)
