Var calculations ? Expousres

[Technically Fundamental]

And people complain about me?

You're being a bit of a nightmare SL mate.

 

[Skill Leverage]

Why, because I won't take someone giving me a hard time, or because his name is in green?

At the end of the day, this is just classic overcomplication. Why do we spend billions of pounds a year on weight-loss products and fad diets? Because we want to believe that it has to be more complicated than 'exercise more, eat less'. It isn't. The same goes for trading - every big local trader I know trades the action, with almost no reference to TA, indicators, var, or anything along those lines. They may well understand them, and they know how the markets can react to them, but they also know that if you're in the business of making money, they're just chaff.

 

[Martinghoul]

SL, just when I thought everything was hunky-dory, you go all bent outta shape... Live and let live, man.

Whatever air of superiority you're detecting here, it is all in your head, so make your own conclusions. As I said a number of times, I am perfectly fine with you doing whatever works for you and I think it would be fair to expect the same of you. Also, I don't work in Canary Wharf, just fyi. I also don't really understand where this rant on what's wrong with the world right now is coming from. I disagree with what you're saying, for the record.

The world doesn't begin and end with 'big local traders'. They know how to make money, but there's other people doing other things that do well for themselves. What is the problem with that?

 

[Technically Fundamental]

I just dont think there is any "air of superiority" as you call.

Also, as I said earlier in the thread I don't believe in VaR myself but what I do realise is that I don't know as much about the markets or trading as the pros. in terms of making money, maybe big locals can rake in as much as institutional traders but you're forgetting that bank traders do thing in a completely different way and have a different business model. It's a different job.

 

[Technically Fundamental]

Anywaaayyyy......

Can I ask some of you professional people using assessing active trades as an example, at what point would you consider using a VaR calculation over just calculating maximum equity at risk by checking stops against openings?

 

[Technically Fundamental]

zactly... now answer my question you massive ego b*stard!!!

 

[Technically Fundamental]

merci

 

[Technically Fundamental]

R:R on individual positions

 

[arabianights]

Aaron, thing is var is based on a normal distribution.

Why is it based on a normal distribution?

Central Limit Theorem.

What does clt assume? Lots and lots of repeated events. That doesn't really apply to any one trade. So it's not really meaningful for your 'risk'...

 

[Technically Fundamental]

Yeh I understand why it wouldnt be usedin just one trade. My question was when would you start to use VaR instead of just calculating exposureby looking at the distance between the stops from either market value or trade entry if running at a loss. That would be easy to set up on an excel. Maybe I'm underestimating the amount of positions held in a portfolio.

Just out of curiousity, how many times have you had real world results that fall into the outlying percentiles when you've used VaR?

 

[arabianights]

I've never used VaR because I am a simple individual with itchy testicles.

The idea behind VaR is not to give a measure of absolute risk (like stops, ignoring slippage, are) but to give an idea of average risk of the entire portfolio.

 

[Technically Fundamental]

I think the accountant in me will never accept the use of averages as a viable tool. Hesus H Christ I hope I get one of these trainee positions I've applied for. You know how much relief it would give me using formulas to caluclate averages over using logic to deduce what some muppet is trying to hide from me and balancing figures to the penny lol.

 

[Technically Fundamental]

from what you're telling me it seems VaR is just another factor to throw in another equation relating to all this cross correlation lark to calculate overall total risk per share or something lol.

Bet it all ends up with a meeting on the current average of the average of the average of all the averages somewhere where some whizz kid will have a board full of thetas and omegas telling some boss he can borrow more money lol.

 

[Technically Fundamental]

Enter KPMG. Earned their fees with that one didnt they lol.

 

[Technically Fundamental]

KPMG does the accounts for most of the banks... they're run into a little bad publicity lately stateside. God knows what they're been getting away with in England with this bunch off backhander swallowing toffs lol.

KPMG sued for $1bn over New Century collapse | Business | guardian.co.uk

 

[Martinghoul]

Aaron, pls feel free to suggest a practical risk management methodology alternative to VaR and let me know... At the moment, you're just being a petulant worthless nihilist, a la Taleb.

 

[Technically Fundamental]

Couldnt give you one mate. I just believe that certain individuals in the sector have shown massive wrecklessness by having these mathematical models that "solve the market" implemented on a global scale. Fair enough you're trying something new so try it small and test it extensively before even considering implementation.

After recent events, I've just lost faith in them that's all. I just think they are worth about as much as the stats that come out of David Camerons mouth.

And who's Taleb?

 

[Technically Fundamental]

This is just like the old mark-to-market vs mark-to-model discussion.

What about showing the maximum plausable (not maximum possible) loss. For positions without stops on say an extreme Major Major S/R level. This could be included in the reports and reasonings people fill in when execution of position trades. Enter it into a database. The open ones would be calculated at the year close and would be shown along with whatever other calculations the bank wants. That way you get to weigh up what could go wrong against what the bank wants you to believe can go wrong. This would propbably be b*stardised too though lol.

I don't know if this is feasible as I'm not familiar with the inner working of a banks but I'm sure something along those lines wouldn't be impossible.

 

[MrGecko]

This is just like the old mark-to-market vs mark-to-model discussion.

No, it's not.

What about showing the maximum plausable (not maximum possible) loss. For positions without stops on say an extreme Major Major S/R level.

This is pretty much what VaR is supposed to do; the problem with it is that it doesn't take into account fire sale conditions when all correlations move to 1 (in the extreme).

If you want to learn about VaR and other techniques, then go away and read the books I suggested.

 

[Technically Fundamental]

Tbh I don't know why I argue about things I dont understand. I think I was more arguing the case against the premise of there being "reliable" financial mathematical models as a whole rather than against VaR itself.