Jump to content

Welcome to the new Traders Laboratory! Please bear with us as we finish the migration over the next few days. If you find any issues, want to leave feedback, get in touch with us, or offer suggestions please post to the Support forum here.

  • Welcome Guests

    Welcome. You are currently viewing the forum as a guest which does not give you access to all the great features at Traders Laboratory such as interacting with members, access to all forums, downloading attachments, and eligibility to win free giveaways. Registration is fast, simple and absolutely free. Create a FREE Traders Laboratory account here.

jperl

Trading with Market Statistics V. Other Entry Points

Recommended Posts

Very good questions thrunner and good observation as well.

First let me clear up one poorly understood idea about the SD. Most traders think that 68.3% of the data falls within 1 SD and 95% falls within 2 SD. This is only true for the normal or gaussian distribution. For skewed data, that is data that deviates from normal behavior, the best estimate can be obtained using Chebysev's inequality, which states that no less than 50% of the data falls within 1.4 SD, and no less than 75% falls within 2 SD. No less than 89% falls within 3 SD. These numbers are quite a bit different than that for the normal distribution.

 

These numbers are of course lower limits. The exact values could be computed from the distibution function. But I don't think there is much to be gained knowing that say 55% of the data rather than 50% of the data fall within 1.4 SD.

 

Another important point which I stated as a theorem in the SD thread, is that for any arbitrary distribution, computing the SD with respect to the VWAP yields the smallest SD possible. What that means in practice is that if you compute the SD with respect to any other price (eg the 1st SD price), you will by the theorem get a larger value for the standard deviation. This implies yet a larger volatility at the 1st SD than it does at the VWAP.

 

These two pieces of information taken together suggest to me that getting to the second SD (computed with respect to the VWAP) is not all that unreasonable although of course with greater risk than trading at the VWAP. Getting to the 3rd SD however is problematic.

 

I will discuss in the next thread, about what to do when you take a trade at the 1st SD and the price action does move against you. There is still room for pulling a profit out of the trade.

 

Recently I decided to understand MP better and immediately thought I should review Jerrys work to see where that fitted and where it didn't. I am gald I did.

 

If I understand this statement correctly taking a daily sample and computing its closing VWAP and SD bands is far more statistically valid for next days trading than using VAH & VAL?

 

Using Chebysev's inequality I wonder what would be a better % to use for MP VAH & VAL to give "no less than xx% gives of the data falls within one standard deviation". VAH & VAL seem to be self fulfilling as S/R however taking into account Chebysev's inequality using MP to detect balance vs imbalance might account for less cases of 'range extension' and more genuine break outs / imbalances?

 

Just a thought

Share this post


Link to post
Share on other sites
Actually ER2 recently has gotten more volatile than even I like. It's getting close to my risk tolerance.

 

 

 

For setups, it doesn't matter what contracts you look at. More important is what the volatility is compared to your risk tolerance. Once you establish a risk tolerance for yourself, and you know what the volatility is of the contract you are trading from the SD, you will know what to trade.

 

Jerry how is ER2 nowadays since it changed to ICE? All of the equity indexes seem to be enjoying a lot of volatility nowadays. Do things work 'better' for you when volatility is slightly less or is it solely about risk tolerance?

 

Risk is simply a function of the width of the bands and the PVP right? Assuming you are well enough capitalised the best markets to trade are those that trade heaviest as there is more statistical data?

 

Do some markets just tend to behave "better" due to other factors (persistence, more consistant volatility etc.)?

 

Cheers.

Share this post


Link to post
Share on other sites
Good observation nelo. I was wondering when someone would see this. The point I will make here, is the old paradigm about choosing trades with a 2:1 or 3:1 reward/risk ratio with fixed stop losses for most traders results in a slow bleeding of their account. We are going to cover this topic in more detail in the next thread, coming soon.

 

 

Probably the most important mind set that I had to overcome, was giving up the idea of fixed stoploss on every trade and substituting the idea of risk tolerance instead. It wasn't until I did that, that I became a profitable trader.

 

 

 

My style is something that developed over many years. I was initially a strong proponent of classical technical analysis and traded futures and stocks for quite a number of years using classical methods. I oscillated back and forth between swing trading and daytrading, but was never satisfied with the results. Some years were profitable, other years were not. There was no consistency. I slowly came to the realization that classical technical analysis was not going to yield a consistent picture of market behavior. It was too heuristic. I wanted day to day consistency. I looked very carefully at market profile analysis. Realized that there was something there but it was woefully incomplete and in some cases just plain wrong. I wanted to be able to write my own software, but there were no good charting packages for doing that until ensign software came along. Being a student of molecular simulation theory, I knew enough about statistics to realize that the logic of the market could be found in a proper statistical analysis of the data. That coupled with understanding risk tolerance and trade management is where I am today. My trading is now quite consistent and I am happy to say has become quite enjoyable. I am both a teacher and student. I've been both my whole life and I am happy to share with you what I've learned about market behavior. There is still much about market behavior that I don't know and learning about the markets will be a lifetime experience.

 

TwMS V (any good approach needs an acronym) has some nuggets bured within it :) Jerry I am fairly confident that I understand your view on risk tolerance however have you tried 'tuning' or does that just produce poorer results.

 

For example putting the stop at the PVP have you tried more than a tick beyond? Maybe 2 or 3 maybe more? How about a stop at the 1 SD beyond the PvP? I know thats even further but if it really boosts your % winners maybe worth it?

 

I know you have talked about the 'never letting a winner turn into a loser' philosophy. As price makes swing highs swing lows does trailing a stop a tick or two behind these pivots hurt profitability much? I think most would sacrifice profit for a smoother equity curve. I gues the question should be does it reduce drawdowns significantly!

 

I know risk and money management are very individual but guess over the years you must have researched various approaches and wondered if you have made observations on the effect of some of these popular methods?

 

Thanks again.

 

Edit: oops maybe this should have gone in the next thread.

Edited by BlowFish

Share this post


Link to post
Share on other sites

If I understand this statement correctly taking a daily sample and computing its closing VWAP and SD bands is far more statistically valid for next days trading than using VAH & VAL?

 

VAH and VAL have no statistical significance. They are purely heuristic values based on a percentage of the volume data above and below the peak. The choice of that percentage was based on the incorrect assumption that the volume distribution is normal. It rarely is. The statistically valid values are the standard deviation of the data from the VWAP. For finite data, this is always definable regardless of the shape of the distribution.

Share this post


Link to post
Share on other sites
Jerry how is ER2 nowadays since it changed to ICE?

 

ER2 hasn't changed to ICE yet but will do so at the expiry of the September contract.

 

All of the equity indexes seem to be enjoying a lot of volatility nowadays. Do things work 'better' for you when volatility is slightly less or is it solely about risk tolerance?

 

I prefer larger volatility. Gives me large profits/trade. Of course more risk too, but still within my risk tolerance.

 

Risk is simply a function of the width of the bands and the PVP right? Assuming you are well enough capitalised the best markets to trade are those that trade heaviest as there is more statistical data?

Risk is determined by what percentage of your capital you can afford to lose on each trade. This is typically 1 or 2 %. You can use the width of the SD bands to determine how far you can expect the market to move against you. If the bands are wider than your risk tolerance then you might not want to trade.

The more data of course the better the statistics. But this does not determine the width of the SD bands. It's the price action that does

 

Do some markets just tend to behave "better" due to other factors (persistence, more consistant volatility etc.)?

 

If by behaving better, you mean not jumping around a lot, I have no idea what determines that, but I really don't care. I just follow the statistics. It's the only thing that really matters. I've quoted this before from Nihabaashi: "To fear volatility is to fear profits"

Share this post


Link to post
Share on other sites

(As an aside, a discussion of distribution skew, also called kurtosis, can be found at this Wikipedia site.

 

I realize I am nitpicking a 2-year-old post, and I apologize for that. But, I'm just now seeing this thread and I have to point out that Skew is not also called Kurtosis. Kurtosis is the 4th moment of a distribution, while Skew is the 3rd.

 

I'm enjoying looking through these threads. Thank you for them.

Share this post


Link to post
Share on other sites
I realize I am nitpicking a 2-year-old post, and I apologize for that. But, I'm just now seeing this thread and I have to point out that Skew is not also called Kurtosis. Kurtosis is the 4th moment of a distribution, while Skew is the 3rd.

 

I'm enjoying looking through these threads. Thank you for them.

 

From Jerrys other posts I have no doubt this was just a slipped gear between brain and fingers :)

 

Enjoy the threads ....it's unusual to come across an approach that is not only novel but has some real value too. Not only that the clarity with which the the ideas are presented is just cream.

Share this post


Link to post
Share on other sites

what if i take a short trade and my risk tolerance is below PVP-stoploss in other words am not able to take the trade...

 

should i calculate my 2% system.-stop in relationship with the PVP-Stoploss when im entering the order?????

Share this post


Link to post
Share on other sites
what if i take a short trade and my risk tolerance is below PVP-stoploss in other words am not able to take the trade...

 

should i calculate my 2% system.-stop in relationship with the PVP-Stoploss when im entering the order?????

 

Forget that one. already found it. (sorry)

 

 

ive been trading vwap's jpearl but... the morning gets really hard to trade... the developing pvp its jumping side to side.... do you guys have some insight to avoid this?. it is a good idea to trade the last hour, when pvp is quiet stable?

Share this post


Link to post
Share on other sites
Forget that one. already found it. (sorry)

 

 

ive been trading vwap's jpearl but... the morning gets really hard to trade... the developing pvp its jumping side to side.... do you guys have some insight to avoid this?. it is a good idea to trade the last hour, when pvp is quiet stable?

 

Ther way I see it there are two approaches to this this issue:

 

You start to trade only after the PVP has stabilized to an extent, so that you know how stable the skew is. It may be after one our or two, but that is something that you have to figure out for yourself - the stability of the PVP or the skew.

 

The other approach is to look at a longer time frame. Which means that the PVP is calculated over many days or a period larger than one day. What this period is something you have to figure out for yourself. This is useful even during periods when there are volume clusters and the PVP tends to swing between them. Taking a longer period stabilizes the skew and the probability of a good trade.

 

Both these aspects were discussed quite well in Jerry(Jperl's) discourse. You should read it more carefully I guess.

 

Jose Kollamkulam

Share this post


Link to post
Share on other sites

I am simming the vWap method. Please look at attached screen shots.

 

1/14/2013 NQ trading.

 

1. PVP was established. Skew was positive. Trade taken at SD1 --> SD2 Long. Profitable trade.

 

2. Price retraced to SD1 again, entered Long according to the Shapiro effect. But price started falling down. Did not scale in at the vWap because there was no Shapiro effect entry.

 

3. Price crashed through PVP, stopped me out for a sizable loss. The skew remained positive all throughout. However, you could notice that a second PVP was building on the top, but never flipped. Is it possible that Ninja dValue indicator is not plotting right? Or if it is plotting right, how could this second trade would've been managed better to avoid the loss?

 

4. Price came back to the vWap. Skew positive. Entered long at the vWap, but price crashed back again. Second loss.

 

How could I have managed these trades better? I wish JPerl posted videos of losing trades as well, not just winning trades. I know, this is too much to ask, and I am very grateful to Jerry for all the input he already provided.

0114vwap_1.JPG.d6c8ceea88d81a44bfe415b4560b7aa8.JPG

0114vwap_2.JPG.377bc0921fd2ddcebd46131ce470de08.JPG

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.