Jesse Livermore - Market Key

Well Hook N. Sinker has this material down cold.

I appreciate immensely his assistance.

I also appreciated all the others who have pondered along this path choosing to comment or not to.

The access to the talent, experience, and intellect on such a platform as this is nothing less than "awesome".

And of course a person such as Jesse Livermore attracts admirers and critics alike for debatable opinions, perspectives and discussion.

I will post a further question for the Elite Traders shortly.

In the meantime Thanks again and a Happy Holiday Season for all!!!
 
Quote from pjones2012:

I like your "set percent" approach. Taking your approach a little further with the index calculations - for example the dow trading in the 8500 range - I would probable test (as you suggest) as follows:

Test A
1. fluctuation parameter = 20%
2. reversal parameter = 10%
More...

Jesse Livermore's method may be undefined at this part of my study. It is not clear to me (1) at what price Mr. Livermore stops a loss, (2) Mr. Livermore's rule for position sizing, and (3) what Mr. Livermore does when he observes a buy signal in his price records but there is no buy signal in the Key value (the sum of prices of each of the two stocks in a group).

Another problem for me is the scaling technique that Jesse Livermore uses. Chapter 6, page 52 in my book Mr. Livermore writes:

"Let us suppose that you want to buy 500 shares of a stock. Start by buying 100 shares. Then if the market advances buy another 100 shares and so on."

I do not know if Mr. Livermore buys another 100 shares every point higher, or every 5 % higher, or exactly what the scaling rule is.

I recall performing trading simulation studies to compare scaling with a single entry, single exit method and overall results were about the same. I do not find great advantage in scaling.

Do you know clearly defined answers to questions 1, 2, or 3 above?
 
Quote from Hook N. Sinker:

Jesse Livermore's method may be undefined at this part of my study. It is not clear to me (1) at what price Mr. Livermore stops a loss, (2) Mr. Livermore's rule for position sizing, and (3) what Mr. Livermore does when he observes a buy signal in his price records but there is no buy signal in the Key value (the sum of prices of each of the two stocks in a group).

Another problem for me is the scaling technique that Jesse Livermore uses. Chapter 6, page 52 in my book Mr. Livermore writes:

"Let us suppose that you want to buy " 500 shares of a stock. Start by buying 100 shares. Then if the market advances buy another 100 shares and so on."

I do not know if Mr. Livermore buys another 100 shares every point higher, or every 5 % higher, or exactly what the scaling rule is.

I recall performing trading simulation studies to compare scaling with a single entry, single exit method and overall results were about the same. I do not find great advantage in scaling.

Do you know clearly defined answers to questions 1, 2, or 3 above?
More...

Hook N. Sinker

Thank you for your post.

O.K.

And thank you for the opportunity to be able to return some favors to you.

First off. Book References

It seems to you read an original 1940 version from Jesse Livermore's book "How To Trade in Stocks" .

First question I have for you is please confirm which version of "How to trade in Stocks" you Reference"?

The version I reference is written by Richard Smitten (Copy Write 2001) where he adds four (04) chapters of his own to make the book twelve (12) chapters rather than eight (08) like I assume your book is.

If the above is your only reference to Jesse Livermore then that is wonderful because that would make you a "purist".

I fortunately, or unfortunately, must bring additional third party reference in order to make the points I have in response to you.

Question 1 - Mr. Livermore's "stop loss":

"The Livermore Money Management System: . . . Rule 2 . . . Never lose more than 10 percent on any trade."

That is it 10 % "stop loss".

He learned this rule as a kid wiping up in the bucket shops. It was the 10% that the bucket shops imposed upon their customers at that time.

Reference: "Trade Like Jesse Livermore" - Richard Smitten (Copy Write 2005) - page X Preface.

Question 2 - Rule for positioning size - Scaling Technique:

Nice question!

I call it pyramiding his position his position. This is what makes Livermore's Method so powerful.

"Money Management Rule 1: Don't Buy Your Entire Position All At One Level". . .
.

"He liked to call this his probe system" .

. . . "For example, if you want to purchase 1000 Shares as the final position do it this way:

Start with a 200 share purchase on the Pivotal Point - if the price goes up, buy an additional 200 shares. Then see how it reacts - if it keeps on rising or corrects and then rises, you can go ahead and purchase the final 400 shares."


Reference: "Trade Like Jesse Livermore" - Richard Smitten (Copy Write 2005) - page 78.

Regarding at what point levels he makes addition purchases (Trades) -

I can provide you with book and page references later - if you desire.

I recall a reference with each "dollar" change in price, he would make additional commitments, when he was in a commodities trade - with one of the grains.

So therefore a long trade:

Trade 1 : price = $50 : 200 shares = 20 %

Trade 2 : price = $51 : 200 shares =20 %

Trade 3 : price = $52 : 200 shares = 20 %

Trade 4 : price = $53 : 400 shares = 40 %

I do not seeing any challenges using $1 for equities trading - in most price ranges.

Question 3: What does Mr. Livermore do when he observes a buy signal in his price records but not a buy signal in his Key Value?

I will take a rain check on this one for now.

I hope this helps in the meantime.













Next - Additional Book References:
 
Quote from pjones2012:

First question I have for you is please confirm which version of "How to trade in Stocks" you Reference"?
More...

My copy of "How To Trade In Stocks" by Jesse Livermore is ISBN 0-934380-20-1 originally published 1940, published May 1991. I count nine chapters in my book. The ninth and last chapter are the "Explanatory Rules".

I remember reading that Richard Smitten interviewed friends and family of Jesse Livermore. Some of these interviews might be the source of additional trading information that appears in some versions of "How To Trade In Stocks" and other books.
 
Livermore used the extreme price of the day. Which would be high or low.
I have written a small book on this and it includes code for the algorithms
<iframe src="http://rcm.amazon.com/e/cm?t=spreadtraders-20&o=1&p=8&l=as1&asins=0975309374&fc1=000000&IS2=1&lt1=_blank&m=amazon&lc1=0000FF&bc1=000000&bg1=FFFFFF&f=ifr" style="width:120px;height:240px;" scrolling="no" marginwidth="0" marginheight="0" frameborder="0"></iframe>
Quote from pjones2012:

I have read a couple great books on Jesse Livermore.

One thing I did not make out (I must have missed it) is what price did Mr. Livermore use for his maket key ananlysis.

What is the: open, high, low, or close.

Does anyone in this forum have any thoughts on this?

Thanks - pjones2012
More...
 
This is an example of the kind of results that I observe when I simulate trading using scaling and non scaling systems:

I use Penney (J.C.) Company (stock symbol JCP) daily closing stock prices adjusted for splits and dividends, 36.90 years from 4 January 1982 to 26 December 2008 in these simulations.

===

Scaling (adding another position) every 5 % price increase, sell when prices decrease 10 %, maximum 5 lots, risk associated with each lot represents 0.5 % of account equity, long positions only, initial capital $ 100,000, 0.5 % slippage and assuming $ 10 commission per transaction.

Profit after subtracting $ 10.00 commission & slippage per transaction: $ 355941
Greatest draw down is 0.0671 (6.71 per cent).
Cumulative Annual Growth Rate (CAGR) is 13.20 per cent.
Annually Compounding Annual Growth Rate (ACAGR) is 5.79 per cent.

===

Without scaling. Same system as above except initial position uses 2.5 % risk and trades single entry, single exit:

Profit after subtracting $ 10.00 commission & slippage per transaction: $ 436809
Greatest draw down is 0.0100 (1.00 per cent).
Cumulative Annual Growth Rate (CAGR) is 16.20 per cent.
Annually Compounding Annual Growth Rate (ACAGR) is 6.43 per cent.

===

I consider these two systems to show about the same performance.

Perhaps Jesse Livermore uses scaling as a technique to keep him trading in the profitable direction. In my copy of "How To Trade In Stocks". chapter 6, page 52, I read "By following this rule you will come nearer being on the right side than by any other method with which I am familiar".
 
A very good discussion can be found on Tradestation's forum:

https://www.tradestation.com/discussions/topic.aspx?topic_id=31001

One of the best diagrams of the Livermore approach is the following attachment.
attachment.php
 

Attachments

Quote from pjones2012:

Question 1 - Mr. Livermore's "stop loss":

"The Livermore Money Management System: . . . Rule 2 . . . Never lose more than 10 percent on any trade."

That is it 10 % "stop loss".

He learned this rule as a kid wiping up in the bucket shops. It was the 10% that the bucket shops imposed upon their customers at that time.
More...

I do not know the percentage of capital Jesse Livermore allocates to an individual trade. A 10 % stop loss policy is convenient for my modeling studies because both fluctuation parameter and reversal parameter can be 10 % under some circumstances. All three parameters can be treated as one parameter.

If I make some assumptions and use the percentage idea that we discussed above then I can write a computer program that simulates Jesse Livermore's Market Key method.

My assumptions are:

1) Buy if price values increase X %. (X is the fluctuation parameter)

2) Sell if price values decrease X %.

3) Position size = (A % of account equity) / (stock price * X %).

4) Single entry, single exit.

5) Long position trades only.

Position size example: Suppose the account shows $ 100,000 equity and I choose to risk 5 % of equity on each trade. 5 % of $ 100000 = $ 5000 bet size. If I use 10 % ( 10/100 = 0.1 if I convert from per cent to fraction ) as the fluctuation parameter and stock price is $ 50 / share then 50 * 0.1 = $ 5 / share. $ 5000 / $ 5 / share = 1000 shares. The position size is 1000 shares in this example.

Can this modified Livermore Market Key actually work?

To answer the first question about trading the Dow Jones Industrial Average (DJIA), the following are trading simulation results using 52.87 years of DJIA index values from 2 January 1948 to 24 January 2007 using a 5 % bet size and $ 100000 initial capital:

Buy if DJIA index value increases 10 %
Sell if DJIA index value decreases 10 %

Total profit 544870
Greatest draw down is 0.1467 (14.67 per cent).
Cumulative Annual Growth Rate (CAGR) is 10.28 per cent.

I can not actually trade an index. Here are the results of a similar study using 13.54 years of daily closing price values for SPY, the Standard And Poors index tracking stock:


Buy if SPY index value increases 10 %
Sell if SPY index value decreases 10 %

Total profit $ 28192
Greatest draw down is 0.1248 (12.48 per cent).
Cumulative Annual Growth Rate (CAGR) is 2.07 per cent.

10 % change for a stock index is big. I try some lesser values:

Buy if SPY index value increases 5 %
Sell if SPY index value decreases 5 %

Total profit $ 109619
Greatest draw down is 0.3325 (33.25 per cent).
Cumulative Annual Growth Rate (CAGR) is 8.04 per cent.

Buy if SPY index value increases 3 %
Sell if SPY index value decreases 3 %

Total profit $ 53221
Greatest draw down is 0.3399 (33.99 per cent).
Cumulative Annual Growth Rate (CAGR) is 3.79 per cent.
 
HNS,what software did you use to backtest scaling??

Thanks in advance

Quote from Hook N. Sinker:

Jesse Livermore's method may be undefined at this part of my study. It is not clear to me (1) at what price Mr. Livermore stops a loss, (2) Mr. Livermore's rule for position sizing, and (3) what Mr. Livermore does when he observes a buy signal in his price records but there is no buy signal in the Key value (the sum of prices of each of the two stocks in a group).

Another problem for me is the scaling technique that Jesse Livermore uses. Chapter 6, page 52 in my book Mr. Livermore writes:

"Let us suppose that you want to buy 500 shares of a stock. Start by buying 100 shares. Then if the market advances buy another 100 shares and so on."

I do not know if Mr. Livermore buys another 100 shares every point higher, or every 5 % higher, or exactly what the scaling rule is.

I recall performing trading simulation studies to compare scaling with a single entry, single exit method and overall results were about the same. I do not find great advantage in scaling.

Do you know clearly defined answers to questions 1, 2, or 3 above?
More...
 
You must log in or register to reply here.
Back
Top