Trading with Market Statistics XI. HUP

[tradepro 10]

Also do you look at value areas and POC in Market profile 5to find a HUP?

[jperl 11]

with the longer term HUP's such as 5,10, and 20 day,do you use a rolling time period or a fixed time?

 

I use the regular trading hours time period for the longer term HUP's although it really is not going to matter that much if you use 24 hour data since the overnight volume is usually not that large.

[jperl 11]

Also do you look at value areas and POC in Market profile 5to find a HUP?

 

I don't use Market Profile.

[pmwhite 10]

First, thanks Jerry for providing such fantastic tools! I've been looking for a way to characterize the market in terms of statistics and this series of threads really cleared up a few things for me. I have a distribution related question - not HUP, but since you're active in this thread I thought I'd give it a shot here as it should be a relatively straightforward question.

 

In a normal distribution you get approx 68% of the observances in the 1st standard deviation. In a developing distribution, if I want to compute the % of observances within the first standard deviation, do I measure it from the VWAP +/- Std Deviation, or from the PVP (POC) +/- Std Deviation? Thanks for your input!

[jperl 11]

First, thanks Jerry for providing such fantastic tools! I've been looking for a way to characterize the market in terms of statistics and this series of threads really cleared up a few things for me. I have a distribution related question - not HUP, but since you're active in this thread I thought I'd give it a shot here as it should be a relatively straightforward question.

 

In a normal distribution you get approx 68% of the observances in the 1st standard deviation. In a developing distribution, if I want to compute the % of observances within the first standard deviation, do I measure it from the VWAP +/- Std Deviation, or from the PVP (POC) +/- Std Deviation? Thanks for your input!

 

You can compute the Std Dev with respect to any starting point, however when computed with respect to the VWAP you can show that this will yield the smallest SD possible. In a normal or symmetric distribution VWAP=PVP so computing the SD with respect to PVP would give you the same SD.

[gasmith7 10]

Jerry, how is your volume distribution function arrived at since you do not use MP? Of course I am referring back to your Trading With Market Statistics I. Volume Histogram; MP is a subset as you say. I have Ensign Windows with it's Price Histogram. How do I need to modify that program? Regards, Gary

[jperl 11]

Jerry, how is your volume distribution function arrived at since you do not use MP? Of course I am referring back to your Trading With Market Statistics I. Volume Histogram; MP is a subset as you say. I have Ensign Windows with it's Price Histogram. How do I need to modify that program? Regards, Gary

 

On the price histogram study window, choose volume. The histogram will then be a volume distribution instead of a price distribution.

[gasmith7 10]

Thank you Jerry, then the time would be one day .... or longer?

[jperl 11]

Thank you Jerry, then the time would be one day .... or longer?

 

You can set the volume histogram for any time period you like. A nice feature is the ability to show the histogram for just the visible bars on your chart.

[gasmith7 10]

Thank you Jerry, the volume histograms on your post of "Trading With Market Statistics I. Volume Histogram" are for that day's volume alone?

[gasmith7 10]

Please reference the above last posting by gasmith7: .... for instance, a one day PVP is obviously much lower than a two day PVP; which would determine how a 'newbie' would trade !

[jperl 11]

Thank you Jerry, the volume histograms on your post of "Trading With Market Statistics I. Volume Histogram" are for that day's volume alone?

 

Yes, that is correct

[Raj 10]

Hi Jerry:

Greetings.

You have mentioned that we can select any number of days to draw the volume histogram including the choice of limiting the data to what is avilable on the current screen.

With this wide choice,, don't the PVP value change depending on volume histogram data ( 1day, or 2 day or current screen data)?

If the PVP value change based on the quantity of data , skew with VWAP also change based on quantity of data used. Given this varience, what is the ideal data period should we use to plot volume histogram and PVP, for day trading the index futures?

 

Appreciate your advise.

 

Regards,

 

Raj

[clmacdougall 11]

Hello Jerry and thank you for all the hard work, I've read and love your posts. Jerry, could you specify the differences between the the VWAP and the VWMA, and why one is better or more useful than the other. Thank you.

[jperl 11]

Hi Jerry:

Greetings.

You have mentioned that we can select any number of days to draw the volume histogram including the choice of limiting the data to what is avilable on the current screen.

With this wide choice,, don't the PVP value change depending on volume histogram data ( 1day, or 2 day or current screen data)?

If the PVP value change based on the quantity of data , skew with VWAP also change based on quantity of data used. Given this varience, what is the ideal data period should we use to plot volume histogram and PVP, for day trading the index futures?

 

Appreciate your advise.

 

Regards,

 

Raj

 

Raj, As a minimun, the amount of data should span a time period at least as long as the period over which you intend to trade. So if you are day trading, then you should have at least one days worth of previous data showing VWAP and volume histogram.

If you trade over a two or three day period, then you should have at least two or three days worth of data.

[jperl 11]

Hello Jerry and thank you for all the hard work, I've read and love your posts. Jerry, could you specify the differences between the the VWAP and the VWMA, and why one is better or more useful than the other. Thank you.

 

The VWMA or volume weighted moving average is like any other moving average except it is weighted by volume. In practice what this means is that you have to choose a time period over which to compute the average. The problem with this, is you are always dropping the oldest data at the back of the average and adding new data at the front.

VWAP on the other hand, does not drop any of the data. It keeps adding data at the front.

Which is better? It depends on your point of view. To me, the VWAP and the histogram from which it is derived provides a clearer picture of the overall statistics of the market. The VWMA and its histogram may have no statistical significance depending on its time frame.

[Raj 10]

Dear Jerry:

 

Thanks for your immediate response.

 

Just to clarify your reply, suppose I start day trading say on Wednesday morning, then I should set my Volume Histogram indicator to plot the summation of Tuesday & wednesday's data as the day progress?

 

Appreciate your advise.

 

Regards,

 

Raj

[BlowFish 10]

Hi Jerry:

Greetings.

You have mentioned that we can select any number of days to draw the volume histogram including the choice of limiting the data to what is avilable on the current screen.

With this wide choice,, don't the PVP value change depending on volume histogram data ( 1day, or 2 day or current screen data)?

If the PVP value change based on the quantity of data , skew with VWAP also change based on quantity of data used. Given this varience, what is the ideal data period should we use to plot volume histogram and PVP, for day trading the index futures?

 

Appreciate your advise.

 

Regards,

 

Raj

 

Raj another way to look at things is from the point of view of the size of moves you want to capture and your risk tolerance. As the distance between VWAP & SD deviation is a measure of volatility and also the minimum size move you might anticipate you can 'tune things' with respect to what you would like to achieve.

 

Edit: hopefully that gives some clues to your follow up question the basic, scalp and swing trading threads have similar characteristics the main difference being the sample size and hence the magnitude (between bands) and the time taken to complete that movement (50 point moves tend to take longer than 5 tick moves after all). Pick what suits you. (I think Jerry mentionde his favourite, can't remember tbh).

Edited May 16, 2010 by BlowFish

[jperl 11]

 

Just to clarify your reply, suppose I start day trading say on Wednesday morning, then I should set my Volume Histogram indicator to plot the summation of Tuesday & wednesday's data as the day progress?

 

 

As a minimun, yes. At the start of Wednesday, you don't have any statistical data to examine. So as a minimum you should have Tuesday's data available and add to it as the new day progresses.

[traderwolf2 10]

Jerry,

 

First of all, thanks for all of your very educational posts. I really appreciate it. Questions on HUP's.......

 

 

Taking out previous day's VWAP and PVP's, wouldn't it be effective to look at previous day(s) volume distributions and look for areas of acceptance and rejection as high probability targets as well?

 

Seems like to me those areas would be the most likely to be HUP's since price has price has some definitive volume history at those places.........

 

So ,,,,,,,,,,,,,,,why not put priority on previous day(s) VWAP, PVP and places of acceptance and rejection as measured by volume?

 

Thanks again,

 

Traderwolf

[traderwolf2 10]

Jperl,

 

Thanks for all of you very educational posts.. Question for you on HUP......

 

 

At the start of a day, I understand to use previous day VWAP and PVP's as HUP's. But would the next best source of HUP's be places in previous days or weeks where price was rejected or accepted as noted by high volume or low volume areas? High volume areas would represent places where price was accepted and low volume areas would be places where price was rejected.

 

I would appreciate your thoughts on this.

 

Traderwolf

[jperl 11]

Jperl,

 

Thanks for all of you very educational posts.. Question for you on HUP......

 

 

At the start of a day, I understand to use previous day VWAP and PVP's as HUP's. But would the next best source of HUP's be places in previous days or weeks where price was rejected or accepted as noted by high volume or low volume areas? High volume areas would represent places where price was accepted and low volume areas would be places where price was rejected.

 

 

Traderwolf

 

Yes of course, you may use any defined HUP's you like, but don't mix apples with oranges. Be consistent. Use the same HUP's from day to day.

[BlowFish 10]

Glad to see you are still around Jerry! Hope life's treating you well.

[jperl 11]

Glad to see you are still around Jerry! Hope life's treating you well.

 

Yes, I'm still here answering the same questions over and over and over....

[Guest feskseabs]

Thank you for the good writeup. It in fact was a amusement account it. Look advanced to far added agreeable from you! However, how could we communicate? I'll just check back at Traders Laboratory - Active Trading Community for you later.