Jack Hershey PVT System Testing (moderated)

[Trader666]

I used your wealthlab code for scoring when I tested buying the "0 to 7 turn" and I've said so before. You all want to claim AFTER THE FACT that my testing should have been confined to a universe of pre-screened stocks, even though Jack's paper said nothing about that. Fine.

But here's a point I haven't made before... all that shows is how flimsy your code and Jack's concept of the price, volume relationship is, because mine doesn't need a pre-screened "universe" of stocks... it works on large, mid and small cap stocks alike without screening. It also tests out favorably in all market conditions, and across multiple parameters.

So thank you all for pointing out so forcefully that Jack's paper on the price, volume relationship is so weak that it needs the crutch of a pre-screened universe to work.

Quote from Spydertrader:

Clearly you must mean "nobody has stepped up to the plate to post code in this thread." Otherwise, nobody would keep refering to the code, I have already posted. So rather than play the moral equivalency game, why not require the code which incorrectly follows the paradigm to be posted, so everyone can see the results.

- Spydertrader

 

[Cocaine]

And yet you are still here. One wonders why you are wasting your time if your system is so awesome you can literally choose ANY stock and make money.

Maybe, just maybe, the method which is in question actually whittles the list down to only the very best candidates poised to make meaningful and catchable moves?

And what is testing "favorably"? 51% success, 99% success?

Quote from Trader666:

because mine doesn't need a pre-screened "universe" of stocks... it works on large, mid and small cap stocks alike without screening. It also tests out favorably in all market conditions, and across multiple parameters.

 

[Spydertrader]

Quote from Trader666:

So thank you all for pointing out so forcefully that Jack's paper on the price, volume relationship is so weak that it needs the crutch of a pre-screened universe to work.

Once again, post your code for all the world to see. Let your 'code' be judged in the light of day.

Quote from Spydertrader:

So why not, rather than once again dig in your high heels, test the final universe stocks with your 'code' and let everyone know how the results turn out.

How many times, Madam, must one ask before you'll simply test and post the code used to test, so everyone can judge for themselves, before you'll stop with this nonesense of who said what, and when did they say it.

 

[NihabaAshi]

Quote from Spydertrader:

Clearly you must mean "nobody has stepped up to the plate to post code in this thread." Otherwise, nobody would keep refering to the code, I have already posted. So rather than play the moral equivalency game, why not require the code which incorrectly follows the paradigm to be posted, so everyone can see the results.

- Spydertrader

First of all, english is not my first language.

Secondly, no moral high game here.

I've asked both sides for the computer code but have only seen those posts ignored by both sides because you guys are too busy mudslinging with each other as documented so far until now.

Thanks for taking a break from the debates to respond.

Further, I thought (starting to think I'm wrong) that the purpose of this thread was for the information to be revealed and questioned in this thread...

Not for people to show up and post links to prior information scatter throughout ET in the past.

Simply, I would like to examine codes by You and Trader666 for doing a comparison.

Now that I have the code by Trader666...

That leaves only you to provide a code and I am not interested in the rule-based method that describes the method as Jack did earlier in this thread.

Therefore, can you say Yes or No on the issue if you have converted Jack's method or your own interpretation of Jack's method from a rule-based method into a computer code...

Yes or No ???

If you don't want to repost your computer code in this thread...

That's ok for me.

By the way, nothing wrong with a rule-based method with in-depth descriptions because I myself use a rule-based method that can't be converted into a computer code.

Edit: Spydertrader thanks for the direct link to Trader666 code (maybe it's Jack's code and the missing code is Trader666)...

http://wl4.wealth-lab.com/cgi-bin/WealthLab.DLL/editsystem?id=51659

Mark

 

[Spydertrader]

Quote from NihabaAshi:

First of all, english is not my first language.

Having read so many of your posts over the years, I would have never guessed English was not your first language. I have always viewed your responses as helpful, and I apologize for the misunderstanding.

Also, the code to which you have linked belongs to me, not trader666. After numerous requests spanning several years she has consistantly refused to post her code.

- Spydertrader

 

[NihabaAshi]

Quote from Spydertrader:

Having read so many of your posts over the years, I would have never guessed English was not your first language. I have always viewed your responses as helpful, and I apologize for the misunderstanding.

Also, the code to which you have linked belongs to me, not trader666. After numerous requests spanning several years she has consistantly refused to post her code.

- Spydertrader

Thanks for the clarification because I was confused once again (my reading skills are slow today).

I now await Trader666 to post his code and if the Anti-Jack folks fail to produce their code...

We then need to question why do they continue with the debates.

Mark

 

[MandelbrotSet]

Enclosed is an excel spreadsheet showing how the code that spydertrader supplied performed for the list of stocks which represent a "Final Universe" according to Jack Hershey's criteria. This is a list of stocks (as posted by ehorn on page 2 of this thread (ehorn's post was taken from theIterative Refinement thread) and the list of stocks was given the "OK" by Jack Hershey on page 3 of this thread.
***
Spydertrader's System Results gave a result of $440,790.53 on the list of stocks, and a Buy and Hold approach gave a result of $803,949.22.

Sounds like Win-Win to me.
***
1. Three stocks could not be charted.
2. One stock gave results which are most likely the result of bad data.
3. I removed the worst performing Buy and Hold stock to compensate for the removal of the Buy and Hold stock with the extreme positve results.

Regards

 

[vikana]

Quote from traderNik:

vikana, this could become a full time job, couldn't it?

: )

It's a pity that there is more interest in mudslinging that actually figuring out the details of PVT. I'm working on first attempt at codifying the simplest aspects of PVT as put forth at the start of this thread.

 

[powerfade]

Quote from vikana:

It's a pity that there is more interest in mudslinging that actually figuring out the details of PVT. I'm working on first attempt at codifying the simplest aspects of PVT as put forth at the start of this thread.

This would actually be of interest, even though the charts I've seen on the SCT/JH threads make me shake my head in wonderment that anyone could make heads or tails of them. Please do let us know what you come up with.

Hopefully you can just delete anything that's remotely off-topic here. This T28/schlap person is particularly useless. Can anyone say OCD?

These 'discussions' look a lot like a religious debate between two parties whose views are written in blood and will never ever change, regardless of any proofs that are provided.

 

[palinuro]

Quote from vikana:

It's a pity that there is more interest in mudslinging that actually figuring out the details of PVT. I'm working on first attempt at codifying the simplest aspects of PVT as put forth at the start of this thread.

The details of the basic version of PVT have been spelled out with great clarity since the second post of Spydertrader's first Equities journal. So it's not as if they have suddenly been revealed.

To backtest the system properly, you need weekly Final Universe lists for the period in question (or a way to derive those lists from historical fundamental data). Using the latest list would probably give reasonably accurate results, since they don't change that quickly, but doing so introduces survivorship bias, plus the posiblity of trading stocks that did not at the time meet the criteria.

Then you need to run a scan on daily data to obtain a list of trading candidates for the next day. Spydertrader's WealthLab chartscripts provide the code for this.

Then you need to process intraday data, using the (beginner) entry criteria Spydertrader provides in his Journal. A complete ATS for this, written for Amibroker, was posted in one of the subsequent Equities Journals.

HTH

-palinuro