StockFetcher Forums · Filter Exchange · historical volatility help<< 1 2 >>Post Follow-up
jhar3
23 posts
msg #51837
Ignore jhar3
5/27/2007 6:15:05 PM

I am trying to write a filter for short term trading, using historical volatility.in the filter i want to show stock that have a historical volatility of 40 percent of better. using (100,6) to express this as a raito, how do i divide the six day readind into the 100 day reading, so that when it drops below 50 percent, it has the potential for a large move as volatility rerverts back to it's mean. thank youfor any help on this filter

alf44
2,025 posts
msg #51860
Ignore alf44
modified
5/28/2007 11:08:42 PM

...that's an old Larry Connors (Street Smarts) technique !

I use this method as part of my current rendition of my Range Contraction/Expansion Filter.


This is how I use it...

----------------------------------

Fetcher[/* Historical Volatility Ratio Calculation (compare a 6 day to a 100 day Historical Volatility) */

set{HistVol6, Historical Volatility(6,1)}
set{HistVol100, Historical Volatility(100,1)}
set{HistVolRatio, HistVol6 / HistVol100}

set{HVRatio, count(HistVolRatio is less than .51, 1)}

Show all stocks where HVRatio is above 0

Average Volume(65) is above 1000000
Price is between 10 and 100

add column HVRatio {vrat.}
]


----------------------------------

This lil filter will show you stocks whose "6 day Historical Volatility"...is 50% or LESS...than it's "100 day Historical Volatility".

In other words...the stock is displaying price CONTRACTION over the last 6 days...relative to the last 100 days !

The HVRatio (vrat.) column shows a "1" where this criteria is TRUE !

I would also add that Connors likes to see an NR4...an NR7...or, an Inside Day...ALONG WITH this Historical Volatility comparison signal...to indicate a possible trade set-up. fwiw

I applied the filter here to ONLY stocks with Avg Volume over 1M and Price between 10 and 100...just to narrow down the choices...you can obviously alter that to your own price/volume criteria if you want.



Regards,

alf44


jhar3
23 posts
msg #51870
Ignore jhar3
modified
5/29/2007 6:50:06 PM

I would to thank you for writing this filter for me. ..I would also ask what nr-4 and nr-7 mean? narrow range? is your Range Contraction/Expansion Filter.on this site, and if so how would I locate it. again thank you very much

alf44
2,025 posts
msg #51871
Ignore alf44
5/29/2007 11:08:32 PM

Glad to help !

NR4 = Narrowist Range of the last 4 days

NR7 = Narrowist Range of the last 7 days

-----------------------------------

Yes...my Range Contraction/Expansion Filter IS posted on this site.

I will "Pop it to the Top"...AND...add "my current rendition" of it.

It has most definitely evolved since I first posted it.


Regards,

alf44


SFMc01
296 posts
msg #51878
Ignore SFMc01
modified
5/30/2007 3:22:48 PM

The "debug filter" indicates that

"and channeltopchange < 0.005
and channelbottomchange < 0.005"

is not recognized. I seem to get the same results with or without it included.


jhar3
23 posts
msg #51943
Ignore jhar3
6/3/2007 4:30:07 PM

alf44, thank's for putting your filter on this web site, alot and thought and work went into it, thanks. I was reading a book from tradehard. com and larry connors has a chapter in it.Quote when you idenitfy markets with readings under 50 percent for three, four or more days, you can rest assured the move will be substantial. what i am going to try is a new filter that looks for both a low 6/100 and a low 10/100 reading on the same day. It is likely that a large move will develop asmost immediately. any help on the 6/100 and 10/100 filter is always welcome

alf44
2,025 posts
msg #51959
Ignore alf44
modified
6/4/2007 10:25:22 AM

jhar3,

Thanks for the nice words about the Range Contraction/Expansion Filter !

-------------------------------------------------------
...you said...

"any help on the 6/100 and 10/100 filter is always welcome"

Take a look at this filter and see if it is what you had in mind :

-------------------------------------------------------

Fetcher[/* Historical Volatility Ratio Calculation (compare a 6 day to a 100 day Historical Volatility AND a 10 day to a 100 day Historical Volatility) */

/* 6/100 Historical Volatility Check */
set{HistVol_6, Historical Volatility(6,1)}
set{HistVol_100A, Historical Volatility(100,1)}
set{HistVolRatioA, HistVol_6 / HistVol_100A}

set{HVRatioA, count(HistVolRatioA is less than .51, 1)}

/* 10/100 Historical Volatility Check */
set{HistVol_10, Historical Volatility(10,1)}
set{HistVol_100B, Historical Volatility(100,1)}
set{HistVolRatioB, HistVol_10 / HistVol_100B}

set{HVRatioB, count(HistVolRatioB is less than .51, 1)}

/* Check for BOTH Historical Volatility criteria to be met */
set{HVRatioAB, HVRatioA + HVRatioB}

Show all stocks where HVRatioAB is above 1

Average Volume(65) is above 1000000
Price is between 10 and 100

add column separator
add column separator
add column HVRatioA {vrat6_100}
add column HVRatioB {vrat10_100}
add column separator
add column HistVolRatioA {vrat610}
add column HistVolRatioB {vrat10100}
add column separator
add column separator
]


------------------------------------------------------

In the above filter...in addition to the 6 day/100 day Historical Volatility check...I ALSO added a 10 day/100 day Historical Volatility check.

THIS filter will ONLY return stocks where BOTH the 6/100 AND 10/100 Historical Volatility Ratios are BELOW 50% (actually 51% or below) !

The columns show a "1" when BOTH Historical Volatility Ratios are BELOW 50% (just as visual confirmation that the filter is returning accurate results).

I ALSO added columns that show the ACTUAL VALUES of the two different Historical Volatility Ratios (again more for confirmation than anything else).

It's kinda kewl I think !


Regards,

alf44



jhar3
23 posts
msg #51991
Ignore jhar3
6/5/2007 5:35:09 PM

alf44, the new filter you wrote works fine. I was rereading dave landry's swing trading ( where this thread started from.) and the one change to your filter that he uses is all of the stocks in his book have a 50 day hv reading of 40% or better. It helps weed out stocks with less potential. I wouldlike to thank you again, If the likes of larry conners and dave landry, jeff cooper and many other top traders use hv in there toolboxes why not us? I look forward to your new filter.

nikoschopen
2,824 posts
msg #52113
Ignore nikoschopen
6/9/2007 7:27:15 PM

Dear Alfie,

Here's a noob question for ya. How is HV any different, in terms of effectiveness, from any other volatility-based momentum oscillators? Obviously, I haven't had much success with many of them. But here's food for thought. Suppose you have ten stocks that you know have very high probability of breaking out. Further suppose that you had to choose only one among those ten stocks. Naturally, you would then want to choose the one that will move the farthest in shortest possible time. I know what I'm saying is unrealistically far-fetched, but how will HV serve any differently from other indicators you would normally use in the selection process?

alf44
2,025 posts
msg #52141
Ignore alf44
modified
6/11/2007 11:46:45 AM

...well, there are so many hypotheticals in this "question(s)" I don't know where to begin.

As for..."how is HV any different, in terms of effectiveness, from any other volatility-based momentum oscillators ?"...

...all I can say to this is that many of the so-called "volatility-based momentum oscillators" that you seem to refer to (Relative Volatility Indicator for example) attempt to measure the DIRECTION of volatility by basically plugging in Standard Deviations of Price Change into some variation of the RSI formula.

Historical Volatility makes no attempt to measure DIRECTION !

Historical Volatility is a measure of past price fluctuations over time. It is a statistical measure of the "realized" volatility of a market (or a stock) in the past. That's it...nothing more !

The classic book by Sheldon Natenberg "Option Volatility and Pricing" defines the calculation of Historical Volatility as...

"the standard deviation of the logarithmic price changes measured at regular intervals of time"

To calculate Historical Volatility, you first find the mean and then calculate the standard deviation. It is then usually expressed as a percentage and computed as the annualized standard deviation of the percentage change in daily price. This is done by multiplying the standard deviation by some factor or time interval...typically this is the square root of 250 (ie. approx. 16) since there are about 250 trading days in a year. This, of course if you're using daily data in your calculations.

As for your "which stock out of ten stocks moves the farthest in the shortest possible time" scenario...I guess if I knew the answer to that question I would be living on my own island right now instead of doing what I'm doing !


Regards,

alf44



StockFetcher Forums · Filter Exchange · historical volatility help<< 1 2 >>Post Follow-up

*** Disclaimer *** StockFetcher.com does not endorse or suggest any of the securities which are returned in any of the searches or filters. They are provided purely for informational and research purposes. StockFetcher.com does not recommend particular securities. StockFetcher.com, Vestyl Software, L.L.C. and involved content providers shall not be liable for any errors or delays in the content, or for any actions taken based on the content.


Copyright 2016 - Vestyl Software L.L.C.Terms of Service | License | Questions or comments? Contact Us
EOD Data sources: DDFPlus & CSI Data Quotes delayed during active market hours. Delay times are at least 15 mins for NASDAQ, 20 mins for NYSE and Amex. Delayed intraday data provided by DDFPlus