RISK

[bullmarket]

good luck in your search then snake

bullmarket

 

[It's Snake Pliskin]

Yup,

Probability!

So in a simplistic case, if you find from your paper trading or whatever simulation technique you use, that given your predetermined entry signals being met (from chart patterns, indicators or whatever) the share price continues to rise on at least the next day 75 times out of a hundred then your entry criteria being met on any single occasion will have a 75% probability of giving a profitable trade for the next day in the future. So in this simplistic example the risk of successful trade is 75% and the risk of a failed trade is 25% imo.

So I can determine from your answer to me that was not addressed to me that you were talking about a totally different topic of probability and paper losses.

 

[bullmarket]

oh dear snake

my post you extracted that quote from was directed at anyone who wants to read it and not specifically at you and was generally in reply to tech/a's M&M's post and had nothing to do with you whatsoever

keep trying though - at least you're entertaining, for me at least

cheers

bullmarket

 

[bullmarket]

anyway, I better go for now - I'll pop back in later this week

good night all

 

[tech/a]

to be consistent then you would also say in my example of the weighted coin coming up heads 600 times out of 1000 that there is still a 50% probability of it coming up heads on any one toss when mathematically that is incorrect and the correct answer is clearly a 60% probability of a head coming up on any one spin.

Hmm would hate to be seen as lacking consistency.

So according to you then I would have a 40% chance of flipping a tail.

So out of 2 possible outcomes(Any one toss as you say) all of a sudden I have a 40% chance of a tail and a 60% chance of a head.

Hahahahaha

Logic also is not one of your strong points!

You've now progressed from annoying to amusing!!
Quicker than I thought possible.

 

[It's Snake Pliskin]

oh dear snake

my post you extracted that quote from was directed at anyone who wants to read it and not specifically at you and was generally in reply to tech/a's M&M's post and had nothing to do with you whatsoever

keep trying though - at least you're entertaining, for me at least

cheers

bullmarket

Well that makes me feel good baby!

anyway, I better go for now - I'll pop back in later this week

Ok you do that. When you come back please elaborate on this:

Calculating risk is by no means a trvial task and so can require extensive number crunching which the average trader would not even consider doing - and I suppose that is why no-one has posted any algorithms for actually calculating risk for a particular scenario

...being your comment, and an investor I am drawing your attention to the words in red.

Snake

 

[lesm]

Risk (n) - exposure to the chance of injury or loss; a hazard or dangerous chance; to run risks, "The Macquarie Concise Dictionary"

Risk
the chance of something happening that will have an impact upon
objectives. It is measured in terms of consequences and likelihood. "Australian Standard - AS/NZS 4360:2004 Risk Management"

Scope of applicability of the standard includes Foreign Exchange and Investments.

 

[bullmarket]

tech/a

try thinking of it this way and it might become clearer - the weighted coin example is similar to a 'loaded' dice.

now in a fair dice all six numbers will have a 1/6 probability of coming up on any one throw, but in a 'loaded' dice at least one of the 6 possible outcomes will have a higher probability of coming up on any one throw....it's as simple as that

good night

bullmarket

 

[Bobby]

tech/a

try thinking of it this way and it might become clearer - the weighted coin example is similar to a 'loaded' dice.

now in a fair dice all six numbers will have a 1/6 probability of coming up on any one throw, but in a 'loaded' dice at least one of the 6 possible outcomes will have a higher probability of coming up on any one throw....it's as simple as that

good night

bullmarket

Well bull??.

Can you clear this up for us.

Coin tossing, that is a fair set of coins =equal, Do you agree that probability theory is concerned with the mathematical analysis of quantities derived from observations of phenomena whose occurrence involves a chance element ?.
What do you think bull of this so far ?.

For a mathematical theory the essential ingredients are that an act or experiment is performed (e.g. coin tossing ) all possible outcomes of which can at least in principle be specified & are observable ( heads or tails ) & that a rule is given whereby for each possible outcome or set of outcomes there can be defined in a consistent way called the probability of the outcome or set of outcomes in question .

Yes or no bull ?

Have Fun Bob.

 

[jurn]

hello,

I've only started buying shares this year so I haven't looked much into risk.
but for me its the shape of your normal distribution bell curve thing at a point of time in the future.

I decided to look into things 1 year from now.
cash in my ING direct account around 5.5%. so my bell curve for this one is really a spike at 5.5%
I decided the blue chip shares I bought would return about 10%, based on past performance.
and some of the smaller ones I bought might return higher, but would be 'riskier', so their bell curve moved to around 20% but are much flatter and have a larger portion to the left side of the peak.
then I've mentally "multiplied" these together based on the proportion of my portfolio to each of these to generate my overall bell curve.


all fun.
Jurn

 

[sails]

.... Risk is not to be confused with probability---Risk is not to be confused with probability.---you get the picture. ...

I have generally thought that risk and probability have some bearing on each other - but it is an interesting concept Tech is trying to get across. Not easy going though as this thread seems to have gone off on a tangent discussing probability Perhaps the probability discussions need their own thread and let Tech get on with the subject of risk?

 

[It's Snake Pliskin]

Bullmarket,

I wrote:

The correlation of holdings, positions should be known to understand the risk being carried. If your positions are highly correlated then you are taking on too much risk. All positions behave alike increasing your drawdowns for example.

You responded:

In my personal case I am starting to look at property trusts that invest in the European commercial, industrial, retail and leisure property markets because from what I am seeing the prospective yields are about the same as trusts I am invested in that invest in US property atm and a little higher than what the LPT's in Australia are averaging atm.

Also, since my number 1 priority is income nowadays I personally don't mind if one goes up a bit while another goes down. Obviously I'd like all my investments to be going up at the same time all the time but that is just not going to happen and I certainly don't want to be spending hours on end doing research and analysis trying to find which sector/region etc is most likely to fire and risk getting it wrong. Mrs bullmarket already thinks I spend too much time on the pc as it is

I (and more importantly mrs bullmarket ) are happy with the returns/income we are getting as they are meeting our objectives and I am reasonably confident that in say 5-10 years time the LPT's I am invested in will be higher in value, everything else being equal - how much? only time will tell but it's not my first priority.

Nothing added to the discussion, initially.

Duc then wrote:

But, again, you are not diversified out of the same two sectors. Different continents..........brings up an interesting component of risk, and this is the Concept of CORRELATION

Financial markets have become increasingly correlated under specific conditions, for an example look no further back than 1998, and Long Term Capital Management.

Correlation, and fat tails go hand in hand, and are an essential component of risk, and risk management.

Correlation is a statistical term, and again is calculated via a statistical significance, usually to two standard deviations, and a concurrent confidence interval.

Fat tails are of course outlier events that exceed two, three, four standard deviations, and should statistically only occur once every million years or so.
Unfortunately, they seem to crop up every couple of years or so.

A robust risk management model must, or it would be eminently sensible to at least entertain the thought of incorporating outliers into your risk model, and generating a strategy that would attempt to minimise the damage, if diversification, suddenly correlated to 1

You responded:

I agree in principle with what you say but now you are starting to talk about the extent of diversification within a portfolio to minimise risk.

Let me refer you to my signature below each post. It clearly states that views I post are suitable for my personal circumstances and so may or may not be suitable for other people.

For me personally, being an income investor, being solely invested in those two sectors makes it easy for me to achieve my investment objectives and still be well within my personal risk profile.

I am diversifying within those two sectors and not beyond atm and am happy with the returns I am getting and I am very comfortable with the level of risk I am carrying at the moment.

And all of the above obviously by no means suggest others should blindly copy what I am doing. It works well for me and obviously may or may not work for others depending on their objectives and risk profiles.

Again nothing added to the discussion and you have ignored the important issue of correlation with regard to diversification.

 

[It's Snake Pliskin]

"Risk has two components:
uncertainty, and
exposure.

If either is not present, there is no risk.

Suppose a man jumps out of an airplane with a parachute on his back. He may be uncertain as to whether or not the chute will open. He is taking risk because he is exposed to that uncertainty. If the chute fails to open, he will suffer personally.

Now suppose the man jumps out of the plane without a parachute on his back. If he is certain to die, he faces no risk. Risk requires exposure and uncertainty.

Good point above.

A common misperception is the notion that the more uncorrelated risks a portfolio is exposed to, the lower that portfolio's overall market risk will be.

Sure there is non-diversifiable or relevant risk to contend with regardless of the diversification.

How does one hedge a portfolio other than to invest/trade in the currency of the market in question?

 

[happytrader]

Hi Tech

Please correct me if I am wrong, but are we talking about managing the 'putting your money where your mouth is type risk? That is managing the the uncertainty and fear of a possible loss?

That the only time I see eyes glaze over and listen to people back peddling even after they've strongly defended their view. The analysis paralysis types.

Cheers
Happytrader

 

[ducati916]

The general topic is starting to fragment slightly, which really is as it should be, as the definition of risk is only the starting point to try and get everyone on the same page, and barring some exceptions, most are in broad agreement as to the definition of risk

We can now address some of the other related topic areas;

Quantifying Risk models
Statistical models (probability, correlation, fat tails & standard deviation)
Non-statistical models (determinism)

Analysis models
Fundamental analysis
Technical analysis
Quantitative analysis
Macro-analysis (economic/political)
Micro-analysis (economic/political)

Risk management models
Stoplosses
Hedging
Diversification
Against the box
Options strategies
Dollar cost averaging
Buy & Hold
VAR
Efficient Frontier

So plenty to be getting on with regarding discussion.
jog on
d998

 

[ducati916]

I suppose we could introduce these concepts now as well, as I see they have already been touched on by happytrader, and I think lesm and they are;

Components of risk
Marketability
Liquidity
Leverage
Information flow
Slippage
Trading costs
Taxes

Systemic risk
Exchange risk
Inflation risk
Interest rate risk
Political risk
Information flow

Non-systemic risk
Psychological risk

jog on
d998

 

[ducati916]

Just to add some further components to;

Non-systemic risk
Psychological risk
Regulatory risk
Random event risk (labour disputes, legal actions, etc.)

jog on
d998

 

[tech/a]

happy.

Please correct me if I am wrong, but are we talking about managing the 'putting your money where your mouth is type risk? That is managing the the uncertainty and fear of a possible loss?

Yes thats the idea I have in mind,it is however necessary to as Duc says get on the same page.

While I dont mind discussion on all or any of Duc's points It could well fragment the main purpose of what I originally had in mind being.

The application of Risk and its Management in trading models.

I will then try to place a practical bent on my posts on the topic refering where appropriate back to a trading example so that we can get a clearer picture of Risk and its Management in our day to day trading.

So on Risk and Probability.
When initially taking a trade we have an uncertainty (Risk) and an unknown probability of success or failure (Probability)

So it makes sence to Quantify that risk.

 

[ducati916]

tech/a

While I dont mind discussion on all or any of Duc's points It could well fragment the main purpose of what I originally had in mind being.

I don't think that it will.
You can see from the multiplicity of opinions, that just the simple definition of risk had some rather disparate views.

If we (as a collective forum) build the model in a logical progression, then whatever model is eventually selected by individuals, they can have a transparent progression, and follow the logic of the construction process.

Teaching how to fish, as opposed to tossing out free fish.
Therefore under the variety of sub-headings we have the different elements that should at least be considered, as will be discovered, not all risk management methodologies can be employed, nor all risk negated.

jog on
d998

 

[tech/a]

Duc.

Not everyone is analytical as you are. I'm not and I dont think you need to be.Infact my view is being overly so can affect your trading.Analysis paralysis.But thats what makes exchanges interesting.

The problem then becomes one of Relevance,here we differ to you everything has relevance and is necessary,to me some is and not all is necessary.

As an example most traders believe that you just have to be right.
Using all sorts of analysis to be as certain as they can that when they put their $$s down that it WILL be a right decision.
Fundamentalists look at all sorts of ratios,pulling apart balance sheets looking for a "Warren Buffet" bargain.
All sorts of scoring tabulation has been designed,newsletters written,weighted expert opinion,you name it its been done all in the false belief that to make $$s you must be correct more often that your incorrect.
Tech analysts put so much rubbish on a chart you have trouble finding the price action.They cross reference oscillators,and draw a zillion lines all looking for confirmation in their analysis.

The good news is that you DONT HAVE TO BE RIGHT MORE OFTEN THAN YOUR WRONG

I found I was far better at getting things wrong than I was at getting them right.
Probability theorists have a burning desire to be right.They cant conceive that you can be wrong--dead wrong and STILL make great $$s.
Its a lot less stressful as well.

Anyway I will continue along a defined path of my discussion of RISK with the desire to keep it relevant to trading.
I'm sure Duc and I will get involved in disagreements,and I will become involved in some of Duc's discussions.
But for those concerned about falling in a never ending discussion representing Quicksand,I'll do my best to keep it relevant.

Finally.This is the way I treat risk.
I have found that after 12 yrs this is the way I like to trade.Its profitable as those that follow T/T can attest to.Its easy,no guess work,takes little time and is fun.
Its not sexy and its boring.
I'm finding it difficult to get the time to reply due to Workplace agreement changes that I have to attend to.Apologies.