Before I begin, let me say it out front that what I am going to present here is based on my trading experience so far. Some are ideas that I picked up from various books/authors, some are the result of battle scares. I am not an expert in this, and I am still learning more.

Also, what put forth herewith is something "should" be done. I have to confess out front that I am not doing it in 100% adherence. But this serve as a good guide line when I initiate a trade, so I just share it out.

Please note that this is going to be limited my knowledge, and I welcome exchange of ideas. We are all here to make money from the market, don't we?

OK, let get started.

Position sizing, is the process of the determining how big or small each of your trading size. It try to answer the question of "How many lots must/can I buy?".

There are a few things we have to take into consideration. The obvious constrains are:
1. your financial muscle
2. your risk tolerance
3. your ability to accept drawdown level of your portfolio
4. Your circumstances (time, reflex, etc)
5. Practical issue of lot size, brokerage etc.


Now let us see how this can be tackled:


1. Portfolio Size

Let looks at your financial muscle. For trading purposes, I refer this to your overall portfolio size. This must be the free money that you can afford to lost/get stuck without affecting your daily life. If you need that money to buy the milk powder for your baby tomorrow, go buy it now and don't trade using that money.

This money must not be too small, otherwise you cannot get the diversification you want (to reduce risk to a level you can accept), or you may have the practical trading issues like min lot size (1 lot), you may have to forgo trading opportunity of stock that has high absolute value (eg: DBS at $14.60 a share), or you may incur too high a brokerage compare to your trading profit.

To me the min portfolio size to trade SGX market only without leverage, is S$25k (I am not into leverage/margin, thus I haven't have much experience on that so I shall leave it to the more experienced to comment the effect of leverage). In our example here Mr X has S$100k, a much better choice. If you don't have S$25k, then you may still benefit from the discussion followed, just that you may have to make some adjustment and accept more risk of ruin along the way or you try accumulate your capital first.

The portfolio size I referred here includes cash position, not just the sum invested in stocks, so as to take up trading opportunity when it appears.



2. Max Nos of Trading Positions Held

This you have to determine yourself. The idea is to balance between diversification and your ability to focus on the stock movement.

If you have too few stocks, the fluctuation in your portfolio size may not to your liking (please note that the fluctuation can both be up and down). If you have too many you may have lost touch of every stock in your portfolio. Mr X hold max 10 stocks. I do not know if this number is good for others, but it is good for Mr X. Please note that this does not mean that he does always has 10 stocks in his portfolio at all time. Once he reaches this numbers of stocks in my portfolio, he decide to trim alway the less optimistic counters to replace it with a more optimistic one; if the new one has more optimistic reading that is. Mr X's friend Mr Y may choose not to add any until any of his existing trades closed. You make our own choice here.

The number your select is something you have to be comfortable with, and it is obviously related to your total portfolio size. Smaller portfolio size usually can afford less number of positions (say 3 or 4 semiautonomous open counters), largest portfolio can afford more, if you chose to.



3. Maximum Percentage of Portfolio a Particular Stock Can Have

The reason here is obvious: it try not to let the whole portfolio's performance drag down by 1 or 2 stocks while trying to optimize the performance when you are right. I think there are literatures on this, but the percentage Mr X allocates here is 30%. You choose whatever you like taking into consideration of the above 2 points.

Notice also this percentage is related to point 2 but not directly, as the position size need not necessary be the same for every trade. This may be the result of pyramid, or average up. I shall leave this out for the time being.



4. Size of Max Stop Loss Acceptable (MaxSLA) to You.

Here is a tricky one. There are a few issues at hand.

A. Percentage
First is to look at the size. It is usually pegged to a percentage of portfolio size. Some say 1%, some say 3%, some say more. The smallest the percentage, your portfolio can withstand longer series of consecutive losses before ruin. But with smaller percentage, you expand your portfolio survival probability, but your may hit by a lot of wig saw if your stop related to each trade is too tight, or you limit the growth of you portfolio if your trade size is too small. If I remember correctly, there is a way to find the optimum %, but for now we shall just say Mr X uses 5%, or S$5k of his $100k Portfolio.

B. Dollar Value
Next look at the acceptable part. Looking at the percentage you may not have any feeling. Let us look the numbers. If you have a portfolio of S$100k, a 5% max stop loss is $5000. That is solid five thousand bucks, you can buy the top end notebook you always want with that, or go for a 2 week tours in US. When the time come and you need to cut with a $5k loss, are you able to do it automatically as if is $50? Or will you paralyze/hesitate and hope that price will move in your favour and wait? Very often, the inaction will eventually let the loss grow further and become so big that the you refuse to cut and wait it out. You waited till someday and give up and cut at $10k loss. Now this S$5k stop and ended as S$10k loss, this S$5k stop is useless.
How to prevent this from happening? Take a close look at the absolute monetary value that make you freeze, make you unable to cut and move on. Each of us has a different flush point where we will not cut when called for. You go through all your previous trade records and review/analyse those trades that you refuse to cut and find your own answer. This is the max value that you should not exceed in order the cut efficiently. (Let us assume Mr X's flush point is S$3k)
Please also that note that this absolute number is not constant. It changes as your circumstances / experience changes.

C. Setting Max Loss Per Position
Taking the above into consideration, between $5k derived from calculation and $3k derive from trading records, it is only logical that Mr X should take the min of the 2 so that it can work. So Mr Trader's X max loss acceptable is $3k.

D. Further consideration: Gaps
Eeek! Gaps (down)!
If you carry your position long enough, you may encounter the problem of gap down. Something really bad happened, and the price just gap down, by passing the stop price you set for yourself. The position suppose to incur only S$3k loss, now is suddenly become $6k in the red, can you cut at $6K?
Gaps do not happen frequently, so this step is your call to see if it is optional. Mr X is conservative, so he round down the figure to S$2k arbitrarily.



5. Now with the (MaxSLA) determined, we look at the entry price and the stop price (determination of stop, or stop price, will be dealt with separately). Let say Mr X is buying Fuyu at $0.555 and through his TA, he set the stop price at $0.525. If he will to be stopped out, he loss $0.03 per share. A $2k MaxSLA will enable him buy 66,666 shares. Now SGX trades are in lot size, so he can buy 66 lots.



6. Wait a minutes, buying 66 lot of Fuyu at $0.555 will cost Mr X $36,630 (brokerage ignored). A max 30% of portfolio size limit per position is S$30,000 (see point 3 above). So in order not to over rely on Fuyu, Mr X further reduce his buy order to 54 lots (30000/555 and round down).



7. Mr X then check if the limit of number of stock held in the portfolio, as mention in point 2 above, before proceed with keying in the order.



There are variations in different stages mention above, but I would save that for now as this posting is getting too long. Comments are welcome.

I am building a database of sort that contains symbols specification. TradeStation has one but of course I cannot use it for my software. Beside the common description other information like min tick value, big point value and contract size etc. are important for system testing.

For example, the contract value for Comex Gold is 100 troy oz. The min tick is US$0.1 for a tick value of US$10. The big point value is US$1.0 for a big point value f US$100.

Without this information, it is impossible for proper evaluation of trading system where $risk per trade, position size, proper P&L calculation and account size all need to be calculated.

I understate MetaStock has a symbol database but do they store such info? If they are can they be used in the language formula for their system testing? I went through their online demo at their website last night but I didn't seems to see them featured.

Take this example:

Assume I were to go short Comex Gold at 421.9 with the stop at 429.8. This work out to be US$790 per contract. If my account size is US$100,000 and my %risk per position is 2% this will be US$2,000. Divide this by US$790, the position size will be 2 contracts (US$2,000/US$790) rounded down since no fractional contract allowed. Factor in round-turn commision of US$30 (US$15/contract) and slippage of 2 ticks each way each contract for $110 cost of execution and it is still within the 2% risk.

Is this type of calculation possible in MetaStock?

Note: I do not use MetaStock therefore the question.

CSK:

As far as I know, no such function. It only has limited options in position sizing (see screen shot below) and has no regard to lot size.



System testing is the weakest area of Metastock. Not only does it lack in position strategies, the entry/exit price fixing method is also handicapped. The use of their formula language, as its name implied, are more for coding the formula.

I understand that external DLL library can be developed as plugin, but that is beyond me so I don't know if that can be done in that rout.

What I do know is, there is a quite sophisticated third party system stimulation called TradeSim, which can use Metastock data to do more. Too bad it is too expensive and I don't have any hands on experience to contribute.

Information about TradeSim can be located here:
http://www.compuvision.com.au/Products.htm

Thanks HC,

I see in the screen shot:

1) % of Available Equity
Is this % of AccountSize for Buy/Sell $Value ($commitment) and not for %Risk. I wonder how this is used for futures contract.

2) Number of units
This will make all entries fixed size? Not adjustable for market condition?

And there is no pyramid entries and partial exit feature?

I browse the TradeSim site. It is an add-on for MetaStock like what Rina Systems does for TradeStation for portfolio testing on multiple symbols. Since an "add-on" there will be limitation imposed by the "parent" program design.

CSK:

It is worse that that.

Yes, this is the % of account size - the 10,000 initial equity as in previous screen shot.

Yes, this option will have fixed entry size throughout the whole system test. No other factors was taken into consideration.

There is an option in the system editor that limit the number of multiple positions the system can have. But this is NOT exactly the same as pyramid. It only add position when the buy signal is triggered again when there are positions already opened.

I think we still can use other position adding strategies, but that we have to code it in the buy trigger, which I think is very terrious and did not even give it a try.

Partial exit? Unheard off in the world of Metastock system testing.

Of course there are work around. One obvious way is to run multiple system tests, each with entry/exit at different conditions, then export those test result into other software (Eg: Excel) to do consolidation. But I am too lazy for that.




There are more limitation in Metastock system testing.

From the screen shot of my previous post, there is a [More] button at the right hand side, clicking that brings up another window with the following 2 tags:



The Commission drop down at the right hand side give only 3 options: 1. Points Per Transaction ($) 2. Point Per Unit ($) 3. Percent Of Transaction (%)



Now prepare to be amused.

The entry/exit execution is limited to Open, High, Low or Close. No other selection is available. (The Realistic Market Prices When this box is checked, any triggered order executes at the open of the next bar. I yet to figure out what difference it has to the the selection of Open price).

There is another option in system design stage where the buy/sell order can be placed as limit order (using codes). But I had yet to know how Metastock handle this conflicting instruction on trade execution.



Seeing so many limitations, I have given up hope and not take Metastock system tester seriously other then a prelimary testing of entry/exit signals. Fine tuning the system using Metastock will give me too much grey hairs and bring me foreward so little.

It just does not justify the risk and reward :).

Thanks again HC.

So not much of "soil survey" and "foundation piling" type of education can be learned then.

I am using version 8, which touted to have a "enhanced system tester" that can run the same system test on multiple counters (or the whole exchange if you wish) at one go. Prior to version 8, Metastock system tester can only test on a single stock. You can imagine the frustration especially with optimization is involved then.

I am not sure what has version 9 improved upon in the system tester, but I got a feeling that there isn't much improvement otherwise the developer will shout all over the place.

Here is a demo of Metastock:

http://www.equis.com/marketing/webinar/metastockdemo.aspx

Hi HC,

Thanks for the updated charts. I am a newbie and having read this simple trading strategy, I do like it and agree to the stop loss at 5%, etc.

I just need your view with regard to how to realign a portfolio that is currently only two stocks for reasons of the past in terms of a number of rounds of averaging down, being stuck at lower levels for months and finally having there value now up to reasonable levels.

Having gone through two similar situations, I am convinced after having paid the price to the market and having to dangerously hold two stocks until value returns, portfolio and long term investing is the way to go (my father exhibited this and grew his portfolio 3 to 4 times over the years through a simple strategy and he had stop losses).

:unsure: The problem I am having is that I have two stocks at ratio of 2:1 (A:S) in dollar value. Obviously, I have to reduce the holdings on A but A has been rising steadily and over the months is now 50% above months ago doldrums. However, B has spurted by 50% and could spurt further. Both have a potential of 20% more increases both from strong analyst recommendations and charting.

From what I have said, it looks like if I want a diversification strategy, I have no choice but to reduce the sizeing of A. I would like to pyramid sell A but how should I do this with the current scenario? I also think S is a long term hold really would like to reduce my holdings in A first. I am quite clear about pyramid buying but need a clearer picture of how to do pyramid selling.

I just need some idea of the strategy to take in terms of how to pyramid sell A in this situation where there is a clear and good upside of 20-30%. Portfolio size is about $250K and I need to realign my portfolio in view of lack of diversification. :unsure:

Thanks....

irrigate:

First of all, I am not an expert in doing portfolio rebalancing. Since you had asked, I just share some of my thoughts here, so as to attract more contribution/comments from others.

I am not sure you are saying you like your father's way or you don't. If you like his way, I believe you as a son asking for detailed instructions from him will be easier. May be later you can share with us here what that instructions are so that more can benefit.

Looking at your message, you seems not comfortable about the proportion of the 2 stocks you have in your portfolio, and you wish to sell A, the one with higher proportion.

Here are some of my thoughts for your consideration:

1. Do you really wish to sell A?

You have to ask yourself: What is the risk and reward at current price of A? Any sell signal (technical or otherwise)? Are you nervous and just wish to lock in profit?

It is not unusual to have different position size in the portfolio. In fact if your entry is based on your initial stop price, it is very likely that you have different position size. Add pyramid, the sizes differ even more. So is this difference in size a problem in the first place?

Generally I don't recommend trimming profitable trades. Even if you are trying to lock in profit via partial exit, it is still better if you have some signals that tell you that the stock may have reach a short term peak, or you have reach certain kind of profit/price/trailing stop target. Selling because that stock had performed so well the it disproportion the portfolio is not a good reason for me to sell.


2. Is there a better use of the sales proceed?

After the sales of A, cash holding in your portfolio will increased. Cash carry no market risk, but it also won't generate any return. Is this what you want? Are you going to use that sales proceed to buy another counter? Does this counter offer better reward/risk ratio than A?


3. How much to sell?

If you had decided that you have to sell A, this will be the next question. The simplest answer I have for this is: sell until you are comfortable with that position size. A peace of mind will help your judgement better later.

What is your comfort level, I don't know. But you may wish to ponder the following:

a. With the balance size of A, what is the trailing stop price? How much does that translate into absolute dollar in term of profit forgo? Is that too much for you to handle?

b. What is your reason for entry at the first place? If you intentionally over-size your position in A initially for a short term ride, then selling that oversize portion is logical when the short term profit/stop is reached.

c. What is the liquidity of A? Any problem selling a portion of A now and balance portion at the target price later?

d. How many time you wish to trim the size of A? You have to balance your risk tolerance, profit potential and commission payable. How about splitting your position in A into short term and mid term positions and sell the short term position first?


4. When to sell?

Depending how uncomfortable you are with your position, and your interpretation of the stock direction. If you are extremely uncomfortable, sell sooner. If it is bearable, wait for your short term sell signal to arrive and sell that short term position first.

Hopes that above helps.

HC: Thanks for the opinion... The reason I am thinking of cutting my position in A is that I do not want to trade with margin as much as possible, having been stuck for over a years in a margin situation - even now... Maybe you can elaborate on the money management expect of trading. I suppose my main discomfort now is that I am trading using margin. However, A seems to have a good upside and it seems reasonable to keep it even with margin and this margin is very low % and will never likely hit a margin call at all.

Also, could you give me a real life example of pyramid selling...

I do not think my father's trading system is anything technical, but I believe he did well in the long run.... He was a remisier and is retired and sitting on his profits. He does not believe in penny stocks, he only buys blue chips and once when we were wondering why he was buying a counter, he said - if the CEO can buy at a certain price recently, why can't he buy at today's lower price. Likely he watches what substantial shareholdrs are doing but they must be the blue chip stocks. FOr example, CEO of Singtel recently bought at $2.60 so using my father's method once it drops to a price like $2.40, he would buy like 10 lots with likelyhood of averagin down as he holds a lot of cash in his portfolio. Of course his ability to hold to a stock is always there. Maybe there's more behind it but I will never be able to find out because he passed on about three years ago and we were surprised by the amount of cash he was holding on to!!!

My mum has another trading system which seemed to keep her happy and she buys when volume is high and there is already a run up in prices and she loves to contra if possible but will hold if there is further upside. Do you call this a simplified "momentum" trading....haha... But she's stuck with ACCS and Brilliant today... However she also holds substantial cash relative to her stock holdings, therefore, she can always hold until the price turns good again.

Seems like having the ability to hold is always the best strategy and being able to wait out is the key. I think I will use my father's method combined with the strategy you proposed what do you think?

irrigate:

My experience in using margin is very limited, I was unable to add anything that is useful. Sorry.

Thanks for updating your father's method, which follow insiders purchase. Since I am more technically inclined (meaning I use technical signal to buy and sell), I have no comment on combining using this piece of information.

I am usually to the point and I know somtimes people don't like it. Therefore, usually I don't want to comment. But in this case, I see a potential time bomb ticking so maybe I can do something good, maybe to plant some good karmic seeds (unfortunately can only do for myself) for the future ripening.

From your writing (which don't seem consistent from one to another but this is not important) you seem not to have a good handle of kindergarten stuff. If this is the case, liquidate all positions, stop trading and learn the basic kindergarten stuff. Then go on to develop your own trading plan. One that you are comfrotable with and where you don't need to ask others for entry/exit guidance. This shoud be taken care of by your trading plan.

If your trading plan is not doing this properly then improve it and improve it and improve it until you are comfrotable with your own workmanship. You enter based on your own entry rules, you exit based on your own exit rules. Then and only then resume trading. If you cannot even trust your own workmanship why should you trust others workmanship. What makes you think others (including mine of course) workmanship are better than yours?

I am not referring to HC or Lion (or anyone else who may contribute). I can sense some discomfort in their replies to you and I am quite sure they understand what I am saying. I can sense (and I think HC and Lion too) from your postings that you have already made up your mind in a certain course of action and they cannot advise or change for reason of legal restriction and compliance.

I can tell you that you are fortunate to have come across people here who by replying this way are acting very professionally. I know there are places where people will rush and jump in giving all sorts of advise to you. You may find them helpful but are they? Really?

I cannot say any more than what HC and Lion have already said. I can just add one more link for you, to try to show you the path of stongest resistance, stongest resistance to change for the better.

Risk Managment by Ed Seykota

You may not like what I have just said now. But one day I hope you may recall that guy who thought he was smart, that guy who said unkind things to you, that guy and the people there actually made a lot of sense.

Okay, I have planted my karmic seeds.

Thanks for the good advice.... Please don't worry about offense as what you said just made sense... Yes, somehow although I know about diversification benefits, I seem to have gone against this "kindergarten" knowledge and ended up with two stocks in my portfolio.... Many thanks to Hc for his questions that set me thinking of the situation.

You have sown very good karma. The article is also timely.