Allocation of capital

[pavilion103]

I have been trading a simulator for 6 months using an imaginary $25,000. When I am going to start live trading, that is the amount I will use.

One issue that hits me straight away is the fact that there is a very limited number of trades that I will be able to take at any one time.

I am considering the best method to pick the best possible stocks to maximise my profits. Having read "Stan Weinstein's Secrets for Profiting in Bull and Bear Markets," I was intruiged by the Sector selection process he uses. For someone like myself starting out (and looking to hold for a few days to a few weeks) is this a good approach? Consider the strength of each sector and then take trades which show the best technical strength in the sectors which show the best technical strength?

I apologise if this seems too obvious but I am not used to doing this.

Also I'm thinking it might be smart to have a trading rule where I sell out of stocks which haven't moved in my direction as soon as I had anticipated to take other better opportunities.

I'd love to hear some opinions/wisdom as to capital allocation as I have very limited knowledge of it.

Cheers,

 

[tech/a]

Do some reading on Fixed fractional Position sizing
You need to know how to position size a trade.
This will put you on the right track so you will not be over or under exposed.

 

[skc]

I am considering the best method to pick the best possible stocks to maximise my profits. Having read "Stan Weinstein's Secrets for Profiting in Bull and Bear Markets," I was intruiged by the Sector selection process he uses. For someone like myself starting out (and looking to hold for a few days to a few weeks) is this a good approach? Consider the strength of each sector and then take trades which show the best technical strength in the sectors which show the best technical strength

While I think it's a decent idea in theory, how do you plan to practically adopt this sector approach? Chart based or fundamental based?

Note also that the Australian market is very top heavy. If you look at the sector indices they will be heavily skewed by the few dominate stocks within them.

Financials - Big 4
Insurance - QBE/IAG
Mining - BHP/RIO
REIT - WDC/SGP
Gold - NCM
Consumer - WOW/WES
Healthcare - CSL
Telecom - TLS
Energy - WPL/STO/ORG/OSH
IT - CPU

This may or may not be a hinderance to you assessing whether a sector has strength or not.

Personally I think a fundamental based screen might work better (think rare earth, potash, uranium 12 months back, biotech etc) - although it will most likely put you into the hot sector where hot money can flow in and out very quickly. You wouldn't mind the flowing in part if you get there early enough, but it's the flowing out part that would give you problems.

 

[pavilion103]

Do some reading on Fixed fractional Position sizing
You need to know how to position size a trade.
This will put you on the right track so you will not be over or under exposed.

Tech, I'm uig fixed fractional at the moment.

If I have a stop loss around 5 or 6% from entry price and am risking $500 per trade (2% of $25,000) then the average position size is approximately $8,000, which would allow me to only hold 3 stocks at a time.
I guess this is why i was tossing up the idea of CFDs, not to increase my risk greatly but to be able to take more positions. Ideally though I would want to use IB because of the costs which is definitely the way I want to go medium to long term.

What are your thoughts on this dilemma?

 

[pavilion103]

While I think it's a decent idea in theory, how do you plan to practically adopt this sector approach? Chart based or fundamental based?

Note also that the Australian market is very top heavy. If you look at the sector indices they will be heavily skewed by the few dominate stocks within them.

Financials - Big 4
Insurance - QBE/IAG
Mining - BHP/RIO
REIT - WDC/SGP
Gold - NCM
Consumer - WOW/WES
Healthcare - CSL
Telecom - TLS
Energy - WPL/STO/ORG/OSH
IT - CPU

This may or may not be a hinderance to you assessing whether a sector has strength or not.

Personally I think a fundamental based screen might work better (think rare earth, potash, uranium 12 months back, biotech etc) - although it will most likely put you into the hot sector where hot money can flow in and out very quickly. You wouldn't mind the flowing in part if you get there early enough, but it's the flowing out part that would give you problems.

Are you saying that the heavy weighting in these sectors is unique to the Aussie market? i.e. this isn't the case in the US?

I was going to use technical indicators for the sector charts, but it seems like it might not be worth it if they are to heavily weighted in this way.

I'm really just trying to get some sort of a checklist compiled at the moment. e.g. 1. check the overall market direction/strength 2. Check the sector 3. Check the weekly chart of individual stock to determine what stage and part of EW the individual stock is in. 4. Do the same for the daily chart. (I am looking to hold for a few days to a few weeks)

Something very basic that makes sure I cover all bases while I am a beginner and prone to overlooking very basic things.

 

[markhocky]

Hi Pavilion,

Another great thread topic if I may say so.

I also came across this problem early on in my system development, and it forced me to rethink my whole approach.

A possible solution might be to measure or grade the trade opportunities that arise based on the signals that are given. As a basic example; if multiple indicators are signalling a buy, this might be more bullish than if only a few are. With this approach, you can then choose more objectively between opportunities that are arising simultaneously. You could even adjust your position sizing according to the 'quality' of the signal.

There are loads of ways to finess this approach, for example:

• you might start putting more weight on signals from an indicator which has been working well, and less on others.
• you could place different weight on the signal depending on the indicator level (e.g. RSI > 80 better than RSI > 70 ?)
• if you keep tabs on open positions, you can even use this approach to compare new opportunities to existing ones to help decide if you should switch.

I realise this is not really an answer, and probably raises more questions than anything. As always, the devil is in the detail. But hoping it might at least give you some more ideas...

Just my thoughts.

Mark.

 

[skc]

Are you saying that the heavy weighting in these sectors is unique to the Aussie market? i.e. this isn't the case in the US?

I was going to use technical indicators for the sector charts, but it seems like it might not be worth it if they are to heavily weighted in this way.

I'm really just trying to get some sort of a checklist compiled at the moment. e.g. 1. check the overall market direction/strength 2. Check the sector 3. Check the weekly chart of individual stock to determine what stage and part of EW the individual stock is in. 4. Do the same for the daily chart. (I am looking to hold for a few days to a few weeks)

Something very basic that makes sure I cover all bases while I am a beginner and prone to overlooking very basic things.

I don't know how it is in the US market, but sectors are top heavy in ASX for sure. I haven't read the book you mentioned but there are plenty of books out there that suggest good ideas that cannot be easily implemented in practice.

May be you can browse the charts of similar stocks rather than the sector index itself. If more of them are showing strength / promising patterns than may be they have better odds of succeeding. But it's obviously up to you to decide how much weight you put on such filter.

On the max no. of position issue - it may not be a bad thing you only have 3-4 positions. On a bad down down all 4 stops are hit and you are down 8%. A few more of the same days and you are gun shy already...

And you need to have 2 sanity checks at all times. 1. No position should be larger than X% of the account. 2. Total leverage should be no larger than Y% of the account - you can find 8 positions with tight stops and next thing you know you have $120K of positions on your $25K account using CFDs. Greek default overnight and you get a margin call in the morning...

The choice of X and Y should be based on the stocks you trade. e.g. Putting 100% on CBA although silly probably won't kill you. Putting 100% on MEO will.

 

[sinner]

+1 for skc last post there, that is the good stuff.

I would add:

"Too few positions to trade", is this really a problem? I don't think so. How many positions do you really need?!

Also, do you really need sector analysis or fundamental screens or anything like that? How about:

If you feel the need to mechanise absolutely everything then I like the Relative Strength Index for this job. Instead of filtering based on sector, or fundamentals or whatever: just take the setups as they come. If you approach a day with more setups than you can take, only long the setup with the strongest RSI and only short the setup with the weakest RSI. Which RSI value to use? Depends, are you looking for short term, medium term or long term strength/weakness? A 20 day RSI isn't going to show you how the stock has performed over the last 6 months.

If you don't feel the need to mechanise absolutely everything then just take the setups as they come and use your brain to filter out setups you don't like.

 

[sammy84]

I don't know how it is in the US market, but sectors are top heavy in ASX for sure. I haven't read the book you mentioned but there are plenty of books out there that suggest good ideas that cannot be easily implemented in practice.

Agreed. Don't give much weight to sector analysis, as I may as well look at a few stocks. The most I'll look at sector charts would be when I'm deciding whether to be more skewed in financials or materials/energy. I find I don't invest in other secotors as they don't seem to trend as well.

 

[sammy84]

Tech, I'm uig fixed fractional at the moment.
I guess this is why i was tossing up the idea of CFDs, not to increase my risk greatly but to be able to take more positions. Ideally though I would want to use IB because of the costs which is definitely the way I want to go medium to long term.

You need to weigh this up against having more profitable positions in the market (assuming you have a positive expectancy). A few profitable positons should easily take care of the increased commissions.

I would use CFDs until your account grew to $100k, then you can get the increased margin account with IB.

 

[AlterEgo]

Tech, I'm uig fixed fractional at the moment.

If I have a stop loss around 5 or 6% from entry price and am risking $500 per trade (2% of $25,000) then the average position size is approximately $8,000, which would allow me to only hold 3 stocks at a time.
I guess this is why i was tossing up the idea of CFDs, not to increase my risk greatly but to be able to take more positions. Ideally though I would want to use IB because of the costs which is definitely the way I want to go medium to long term.

What are your thoughts on this dilemma?

2% is just the maximum, it can be lower, and IMO should be lower - 2% is too high IMO. If you lower it to 1% for example, that will give you an average position size of $4000 and you can then hold 6 stocks at a time. But beware of going too low on your position sizing as that makes your brokerage a larger % of your position, meaning you need a larger % gain just to break even on the trade. Ideally you'd want to be trading a larger account size.

 

[tech/a]

Firstly you need to understand fixed fractional position sizing.

As pointed out the % risked should vary on depending on market and in particular stock performance.
Lastly I use margin at every opportunity and I use IB.

Let's look first at CORRECT Fixed Fractional position sizing.
If stock is $10 and you are risking $500 and your stop is 20 c away from your purchase price then you can buy 2500,
You could maintain your exposure by adding margin---to whatever level your portfolio can stand--- read about Portfolio Heat.I keep mine ( heat ) below 20 %
if a stock is flying I am happy to risk 5 % or more but this will normally be for a short term " Rape and Escape" type trade/s.

The answer to type of method used for stock selection CANNOT be answered without proven test results.To attempt to is just guessing.
Even Wienstien's method is taken on face value --- unless you've tested it and you KNOW your application has positive expectancy over wide and varied markets and Bourses.
Wienstien may not be happy with the result!!!!!

Lastly all methods work brilliantly in bull markets.
Select your method and testing to suit the market--- if you expect all methods to work in ALL markets you'll be continually pulling your hair out and losing your capital by slow bleeding---- if your good----- or gushing wounds---- if your crap.

 

[pavilion103]

+1 for skc last post there, that is the good stuff.

I would add:

"Too few positions to trade", is this really a problem? I don't think so. How many positions do you really need?!

Also, do you really need sector analysis or fundamental screens or anything like that? How about:

If you feel the need to mechanise absolutely everything then I like the Relative Strength Index for this job. Instead of filtering based on sector, or fundamentals or whatever: just take the setups as they come. If you approach a day with more setups than you can take, only long the setup with the strongest RSI and only short the setup with the weakest RSI. Which RSI value to use? Depends, are you looking for short term, medium term or long term strength/weakness? A 20 day RSI isn't going to show you how the stock has performed over the last 6 months.

If you don't feel the need to mechanise absolutely everything then just take the setups as they come and use your brain to filter out setups you don't like.

Thanks for the reply.

I don't want to use fundamental analysis at all for the sectors. I read a bit about Relative Strength in the same book I was reading. This must be different to RSI. It basically says that if XYZ stock rises by 10% but the market average advances by 20% then obviously XYZ is relatively weaker.
It uses the formula: Price of XYZ/Price of Market Average
The example it uses is 50/310 = 0.16
(This is on page 108 for anyone who has the book)

This has confused me. Can you or someone else please explain what relative strength indicators they use and what they mean exactly?

 

[pavilion103]

Firstly you need to understand fixed fractional position sizing.

As pointed out the % risked should vary on depending on market and in particular stock performance.
Lastly I use margin at every opportunity and I use IB.

Let's look first at CORRECT Fixed Fractional position sizing.
If stock is $10 and you are risking $500 and your stop is 20 c away from your purchase price then you can buy 2500,
You could maintain your exposure by adding margin---to whatever level your portfolio can stand--- read about Portfolio Heat.I keep mine ( heat ) below 20 %
if a stock is flying I am happy to risk 5 % or more but this will normally be for a short term " Rape and Escape" type trade/s.

The answer to type of method used for stock selection CANNOT be answered without proven test results.To attempt to is just guessing.
Even Wienstien's method is taken on face value --- unless you've tested it and you KNOW your application has positive expectancy over wide and varied markets and Bourses.
Wienstien may not be happy with the result!!!!!

Lastly all methods work brilliantly in bull markets.
Select your method and testing to suit the market--- if you expect all methods to work in ALL markets you'll be continually pulling your hair out and losing your capital by slow bleeding---- if your good----- or gushing wounds---- if your crap.

Thanks Tech,

I'm not sure if I am not understanding you (please be patient if I am) or if you misunderstood what I was trying to communicate with the fixed fraction positioning but that IS the way I have been doing it. I mean, I have been risking $500 per trade and calculating my position size as $500/(distance between entry and stop) e.g. $500/0.20 = $2,500 as in your example. Is this correct? Did we just get our wires crossed?

I hadn't considered using different levels of risk for different stocks. I guess this is something for me to look at as I become more experienced in determining what an incredibly strong trading opportunity is. Would that mean that potentially I could have categories/rankings for trading opportunities e.g. A+, A, A-, B+ and then allocate a greater % risk to the better opportunities? Or is it better to stick with a fixed 1% or 2% and then only risk more for the occasional strong trade? (of course I will need to test this myself, but wouldn't mind an opinion if you choose to answer this).

Lastly, the reason I said CFDs (which I don't want to do) is because I'm not sure if it's possible to use margin on a $25,000 account which I will start with. Can anyone help confirm if this is true or not? I'll take a look myself on the website.

Thanks

 

[AlterEgo]

Thanks for the reply.

I don't want to use fundamental analysis at all for the sectors. I read a bit about Relative Strength in the same book I was reading. This must be different to RSI. It basically says that if XYZ stock rises by 10% but the market average advances by 20% then obviously XYZ is relatively weaker.
It uses the formula: Price of XYZ/Price of Market Average
The example it uses is 50/310 = 0.16
(This is on page 108 for anyone who has the book)

This has confused me. Can you or someone else please explain what relative strength indicators they use and what they mean exactly?

Yes, it can be confusing, because despite it's name, RSI has absolutely nothing to do with relative strength. The "traditional method" of calculating relative strength (called the "ratio method" in The Encyclopedia of Technical Market Indicators) is comparing the performance of one security against another. As per your example above, you divide the stock price by the index (or whatever else you want to compare it against) and plot the result as an indicator. If this indicator is rising, it means the stock is outperforming the index. If the line is falling, it's underperforming the index.

This book also talks about the "screening method" (which is basicly what I do), which runs an exploration over the entire universe of stocks and ranks them from best to worst performer. William O'Neil uses this type of method in his book.

 

[Julia]

I hadn't considered using different levels of risk for different stocks. I guess this is something for me to look at as I become more experienced in determining what an incredibly strong trading opportunity is.

Are you perhaps over-complicating your approach? If you really haven't considered that different stocks will have considerably different risk profiles, then you're surely ignoring something that's absolutely basic.

Do you think your capital will be equally safe in e.g. a speccie miner with no proven track record as it will be in e.g. one of Australia's top companies with a proven record of growth and returns to shareholders?

 

[tech/a]

Yes
IB will give you 2:1 over night
And 4:1 intraday.on a 25 k account.

 

[tech/a]

Yes
IB will give you 2:1 over night
And 4:1 intraday.on a 25 k account.

The point I was making and failing to communicate well was that if say you had a $10 stock with a 10c risk and a $500 Capital allocation you could buy 5000 which is ove your total investment capital.
You an and should use margin to leverage the position.
You are still risking only a the same % as any other trade-- using he same position sizing method.

I trade---currently smalls with a buy price of less than $1 and more than .05 c
This is where the bang for buck is in my view.

 

[skc]

The point I was making and failing to communicate well was that if say you had a $10 stock with a 10c risk and a $500 Capital allocation you could buy 5000 which is ove your total investment capital.
You an and should use margin to leverage the position.
You are still risking only a the same % as any other trade-- using he same position sizing method.

I trade---currently smalls with a buy price of less than $1 and more than .05 c
This is where the bang for buck is in my view.

Yes and that $10 stock goes into a trading halt and comes out with a profit downgrade, a capital raising and 30% lower share price. The account is down 60% despite the 10c initial risk.

It's up to you how much you want to avoid this kind of situation. And trust me if you trade long enough you will run into it.

Just as likely as your $50K position enjoys a takeover with a 30% premium, mind you.

 

[sinner]

Yes, it can be confusing, because despite it's name, RSI has absolutely nothing to do with relative strength. The "traditional method" of calculating relative strength (called the "ratio method" in The Encyclopedia of Technical Market Indicators) is comparing the performance of one security against another. As per your example above, you divide the stock price by the index (or whatever else you want to compare it against) and plot the result as an indicator. If this indicator is rising, it means the stock is outperforming the index. If the line is falling, it's underperforming the index.

This book also talks about the "screening method" (which is basicly what I do), which runs an exploration over the entire universe of stocks and ranks them from best to worst performer. William O'Neil uses this type of method in his book.

It's only confusing if you confuse yourself.

Relative Strength is as you have both described, but not what I referred to, even once in my post.

I am referring specifically to the Relative Strength Index (as stated), which measures the strength of any given bar relative to a prior range of bars.