Position size advice for first time investor

[MrEclectic]

Hello everybody,

I'm a first time investor, hoping to put about $20,000 into various shares and then continue to add more contributions to my portfolio as time goes by.

I've tried to read up on various investing strategies, but I've become a little confused by position size as suggested by Colin Nicholson. While the concept of avoiding investing more than a certain amount of capital in any one stock seems sensible enough, I'm struggling to comprehend the numbers he's using to calculate the maximum investment per company.

Does anyone have any recommended resources to learn more about position size? Incidentally I'm not wedded to Nicholson's calculations by any means (particularly as his portfolio size is much, much larger than my modest start), so any strategies or explanations will be helpful!

 

[tech/a]

Hello everybody,

I'm a first time investor, hoping to put about $20,000 into various shares and then continue to add more contributions to my portfolio as time goes by.

I've tried to read up on various investing strategies, but I've become a little confused by position size as suggested by Colin Nicholson. While the concept of avoiding investing more than a certain amount of capital in any one stock seems sensible enough, I'm struggling to comprehend the numbers he's using to calculate the maximum investment per company.

Does anyone have any recommended resources to learn more about position size? Incidentally I'm not wedded to Nicholson's calculations by any means (particularly as his portfolio size is much, much larger than my modest start), so any strategies or explanations will be helpful!

Google Fixed Fractional Position Sizing

And adjust to your requirements

 

[So_Cynical]

Position size for buy and hold investing = how much are you comfortable losing?

 

[Ariyahn2011]

I do about 2000 each company over a 20K investment. I have no stop loss. I probably should. I only sell when I believe the company is flooked and the management is poor. Then i sell. I was told Warren Buffet doesnt even look at the share price. Is that right?
Cheers

 

[Julia]

I do about 2000 each company over a 20K investment. I have no stop loss. I probably should. I only sell when I believe the company is flooked and the management is poor. Then i sell. I was told Warren Buffet doesnt even look at the share price. Is that right?
Cheers

I'd have no idea what Warren Buffet does other than buy companies which he believes are undervalued then putting his own management in to turn them around.
That is somewhat different to buying shares in some company and declining to observe the SP.

Why do you think it would be useful not to watch what your investment is doing?

 

[Boggo]

... I have no stop loss. I probably should. I only sell when I believe the company is flooked and the management is poor. Then i sell. I was told Warren Buffet doesnt even look at the share price. Is that right?
Cheers

It would probably not achieve much to state the obvious, we have all had our first month of our ten year apprenticeship that we look back on.

I'd have no idea what Warren Buffet does other than buy companies which he believes are undervalued then putting his own management in to turn them around.
That is somewhat different to buying shares in some company and declining to observe the SP.

Exactly Julia.

 

[Funda-Struck]

Hey Mr Eclectic,

I have been using my stop loss level and amount of money to risk to determine my position size.
If you get hold of Nick Radge's book "Adaptive Analysis" he can explain it better, but basically it is the risk amount divided by the stop loss amount equals number of shares to buy.

Probably won't work as well with 20K, so I'll use a $100K example.
You may choose to risk 1% of your capital on a trade, $1000. (1% of $100, 000).
The Share price is $1.00, the stop loss is set at 90c, equals a 10c gap.

$1000 divided by 0.10 equals 10000.
Therefore your position size would be 10000 shares at $1 each, $10,000.
Basically this means that if the price drops triggering your stop you are limiting your loss to $1000.

...I guess this set up is more for a trader though, not a long term investor. (also this works for me, everybody else may have a different opinion!)

Good luck mate.

 

[tech/a]

Hey Mr Eclectic,
I have been using my stop loss level and amount of money to risk to determine my position size.
If you get hold of Nick Radge's book "Adaptive Analysis" he can explain it better, but basically it is the risk amount divided by the stop loss amount equals number of shares to buy.
Probably won't work as well with 20K, so I'll use a $100K example.
You may choose to risk 1% of your capital on a trade, $1000. (1% of $100, 000).
The Share price is $1.00, the stop loss is set at 90c, equals a 10c gap.
$1000 divided by 0.10 equals 10000.
Therefore your position size would be 10000 shares at $1 each, $10,000.
Basically this means that if the price drops triggering your stop you are limiting your loss to $1000.

...I guess this set up is more for a trader though, not a long term investor. (also this works for me, everybody else may have a different opinion!)

Good luck mate.

This is Fixed Fractional Position Sizing

You can trade without a stop and as has been stated just trade a % of your total equity.
The thinner you spread it----the more positions you have---- the less risk you are taking on.
But there are other risks like portfolio risk---- having too many stocks in your portfolio in one sector----eg mining.

With larger balances you can combine both Fixed Fractional and % of total equity.

If your trading Margin CFD's you must learn how to use leverage properly with good risk management.
Margin/leverage can be used without exposing you any more than using your initial balance.

Learn how to!!

 

[dead trader]

If your trading Margin CFD's you must learn how to use leverage properly with good risk management.
Margin/leverage can be used without exposing you any more than using your initial balance.

Learn how to!!

Hey tech/a,

How do I learn? Are there any good books, or should I just google?

Thanks in advance.

 

[MrEclectic]

Hey Mr Eclectic,

I have been using my stop loss level and amount of money to risk to determine my position size.
If you get hold of Nick Radge's book "Adaptive Analysis" he can explain it better, but basically it is the risk amount divided by the stop loss amount equals number of shares to buy.

Probably won't work as well with 20K, so I'll use a $100K example.
You may choose to risk 1% of your capital on a trade, $1000. (1% of $100, 000).
The Share price is $1.00, the stop loss is set at 90c, equals a 10c gap.

$1000 divided by 0.10 equals 10000.
Therefore your position size would be 10000 shares at $1 each, $10,000.
Basically this means that if the price drops triggering your stop you are limiting your loss to $1000.

...I guess this set up is more for a trader though, not a long term investor. (also this works for me, everybody else may have a different opinion!)

Good luck mate.

Thanks for the advice from yourself and the rest of the commenters. It looks like the ASX is not off to a great start this year so I have some time to do a little more research, but your explanation is much clearer than the one Nicholson gives in his book!

 

[The Bear]

Hey tech/a,

How do I learn? Are there any good books, or should I just google?

Thanks in advance.

A couple of free websites that will help you out.

http://www.learncfds.com/CFD-Tutorial-Introduction.html#.VK4G4HsXt8F

http://www.contracts-for-difference.com/risks/CFD-leverage.html

http://www.investopedia.com/articles/stocks/09/trade-a-cfd.asp

 

[howardbandy]

Greetings --

Essentially all of the position sizing literature assumes the distribution of trades is stationary. It is not.

Think through how you will determine whether your system is healthy or not. System failures begin with small drawdowns. Drawdowns are warning to reduce position size.

Proper position size depends on the current degree of synchronization between the model (logic, rules, and parameters) and the data. Position size cannot be determined from within the system. It can only be determined by analysis of trades as a separate process. If position sizing is put within the system, that removes the "knob" that is turned to manage the health of the system.

Best,
Howard

 

[tech/a]

A couple of free websites that will help you out.

http://www.learncfds.com/CFD-Tutorial-Introduction.html#.VK4G4HsXt8F

http://www.contracts-for-difference.com/risks/CFD-leverage.html

http://www.investopedia.com/articles/stocks/09/trade-a-cfd.asp

How do they help
None show position sizing correctly using leverage.

Howard how can you test a system without nominating a position size?
How can we alter it relative to results being achieved and how do we find the optimum?

 

[The Bear]

How do they help
None show position sizing correctly using leverage.

Howard how can you test a system without nominating a position size?
How can we alter it relative to results being achieved and how do we find the optimum?

Hi Tech,

Yep you're correct had a more involved looked and it doesn't actually explain how to position size with leverage.

More just some general info on CFD's and Leverage.

Although I'm only new to this game and yet to trade, let alone try CFDs from In the Red example about using Fixed Fractional Positioning he worked out we could buy 10,000 share @ $1.00 = $10,000 worth of shares.

My understanding (correct me if I'm wrong) is if I was using a 10% margin and a $20 brokerage fee, my total amount of shares to be purchased to allow for this 10% margin would be.

[(Buy Price $1.00 * Number of shares purchased 10,000) * (100% - 10%)] + Brokerage $20 = $9,020

If the margin was 90% then it would be:

[(Buy Price $1.00 * Number of shares purchased 10,000) * (100% - 90%)] + Brokerage $20 = $1,020

That's my guess anyway :/

 

[Funda-Struck]

Just my opinion bear, but if you are fairly new to shares stay clear of cfds for now. Probably play with normal equities first to test and develop your strategies.

 

[The Bear]

Just my opinion bear, but if you are fairly new to shares stay clear of cfds for now. Probably play with normal equities first to test and develop your strategies.

Hi In The Red, don't worry that's exactly what I intend to do, won't be even looking at CFDs until I'm profitable with shares. Was just commenting to extend my knowledge and learn from more experienced players on ASF.

 

[dead trader]

Hello ITR, I'm also new to CFDs and trading live but having good luck, shall I migrate to shares or stay with the CFDs?

 

[howardbandy]

How do they help
None show position sizing correctly using leverage.

Howard how can you test a system without nominating a position size?
How can we alter it relative to results being achieved and how do we find the optimum?

Hi tech/a --

My recommendation is to test using equal-sized trades. For stocks, that is equal dollar amounts. Say $10,000 for each trade. For futures, use one contract. Ideally, the objective function used to rank results should be independent of metrics that are sequence dependent. Good metrics are the ratio of winners to losers, average percent gain per trade, etc. Maximum drawdown is a poor metric. Maximum equity when compounding or position sizing is included is a really bad metric -- but it is often the one that is the default for the platform.

The idea is that the model (logic, rules, and parameters) identifies patterns in the data that precede profitable trades. During development, the model is repeatedly modified and retested until satisfactory results are found. This is in-sample.

For trading systems, in-sample results alone have no value in predicting future performance. We need to see what happens when the model is shown data that has not been used in development -- out-of-sample data.

If the system is sound, the tuned model will identify profitable trades in the out-of-sample data. The best we can hope for is that the distribution of trades continues in the future as it did in the past. What we cannot count on is that the same trades will occur in the same order. The estimate of future performance comes from a Monte Carlo analysis of the trades taken in different order. That is a separate step. That process is the main focus of my "Modeling" book.

My next book, "Quantitative Technical Analysis," goes into much more detail, beginning with an analysis of the risk inherent in the data series (before any model is applied to it), the risk tolerance of individual traders, the risk associated with a trading system (model plus data), the determination of the maximum safe position size, and the estimation of the profit potential. It also introduces machine learning for trading system development -- but how the system was developed is a side issue when the discussion is risk.

Best regards,
Howard

 

[tech/a]

Thanks understood.

When is the new book due for release?

 

[howardbandy]

Thanks understood.

When is the new book due for release?

I am finishing up some changes and additions suggested by an editor. And making certain the table of contents and index are correct. I expect to be done within a week. The book will be available shortly after that.

We are moving all of our order processing and fulfillment to Amazon. I will post some messages when the book has been received into the Amazon distribution system.

Best,
Howard