What will HowardCohodas's month on month return be for March?

[Mr. Charts]

no sir.

You could do the same with a toin cossing account. Bet on heads and withdraw everytime you're up X amount. Doesn't make it profitable.

Is that HoCo's version?

 

[Mr. Charts]

HoCo,
This travel agency business of yours, is it going anywhere?
(All meant in good humour, Howard )

You should link up with HoJo's. I can just see it now, a HoCo's travel franchise in every HoJo's.

For those wishing to travel but also to stay at home, the HoCo travelator with added Calorie Control.

Better prospects than the trading ideas methinks

With your irrepressible, bubbly, I can do no wrong personality I'm sure retirement will be a fascinating experience; just don't let all those Black Swans eat you out of house and home - they do have their beady eyes on you


Richard

 

[tar]

no sir.

You could do the same with a toin cossing account. Bet on heads and withdraw everytime you're up X amount. Doesn't make it profitable.

Regarding MM : Trading is about making money not making points and pips , so i would say MM plays a big role ( the biggest ) in a successful trading strategy , at the end of the day you could be down 10 points/pips but still made money ...

 

[tar]

Is that HoCo's version?

no it is yours

 

[HowardCohodas]

HoCo,
This travel agency business of yours, is it going anywhere?
(All meant in good humour, Howard )

You should link up with HoJo's. I can just see it now, a HoCo's travel franchise in every HoJo's.

For those wishing to travel but also to stay at home, the HoCo travelator with added Calorie Control.

Better prospects than the trading ideas methinks

With your irrepressible bubbly I can do no wrong personality I'm sure retirement will be fascinating


Richard

I get travel agency commission automatic deposits. Does that count?

I am retired. Haven't been so busy or had as much fun in years.

 

[rosehound]

Howard,

Feel free to repost, but please post any commentary/riposte you may attach to my comments in this thread as well - might as well try and continue a dialog in an open thread.

Cheers

 

[HowardCohodas]

Howard,

What component of your trading system do you believe gives you an 'edge*'?

*A trading edge is a unique skill/strategy that differentiates a trader from other traders in that the said trader is able to consistently rake in profits.

I've always been troubled by this definition of trading edge and have had difficulty fitting a discussion of my strategy into it. When you brought it up again in the attached post, I thought I'd do some research into other attempts to define it.

The one I found that seems the simplest and most elegant is positive expectancy. Would it be satisfactory if I described my strategy in those terms?

 

[rosehound]

I've always been troubled by this definition of trading edge and have had difficulty fitting a discussion of my strategy into it. When you brought it up again in the attached post, I thought I'd do some research into other attempts to define it.

The one I found that seems the simplest and most elegant is positive expectancy. Would it be satisfactory if I described my strategy in those terms?

Howard,

This is the absolute crux of the problem, my previous post is saying (and indeed other critics of your strategy all agree) that your system has negative expectancy.

 

[robster970]

HoCo will demonstrate that his system has a +ve expectancy based upon his current result set which the others know is not a significant enough sample size to highlight said negative expectancy for this type of system.

Attempts to discuss this qualitatively through demonstrating that the cause (or edge) that generates the +ve expectancy is neither found in a directional bias nor a mispricing of the options and is purely down to a statistical anomaly associated with a high win rate system are lost on HoCo because he refers back to his statistically insignificant sample showing +ve expectation to say that his edge is somewhere else, he doesn't know where and neither do his detractors.

 

[the hare]

HoCo will demonstrate that his system has a +ve expectancy based upon his current result set which the others know is not a significant enough sample size to highlight said negative expectancy for this type of system.

Attempts to discuss this qualitatively through demonstrating that the cause (or edge) that generates the +ve expectancy is neither found in a directional bias nor a mispricing of the options and is purely down to a statistical anomaly associated with a high win rate system are lost on HoCo because he refers back to his statistically insignificant sample showing +ve expectation to say that his edge is somewhere else, he doesn't know where and neither do his detractors.

At last ! This sums up the situation perfectly.

 

[HowardCohodas]

HoCo will demonstrate that his system has a +ve expectancy based upon his current result set which the others know is not a significant enough sample size to highlight said negative expectancy for this type of system.

What sample size would be sufficient?

 

[rosehound]

What sample size would be sufficient?

Several years at least, although with your system I (and many others) believe that it is possible to determine the results/expectancy of this system a priori, and therefore technically I probably think that sample size is irrelevant.

 

[robster970]

What sample size would be sufficient?

Where do you think your edge is that causes +ve expectation? That might give people a hint to help you figure out what a good sample size may be. Money management does not in itself give you +ve expectation btw.

 

[HowardCohodas]

Several years at least, although with your system I (and many others) believe that it is possible to determine the results/expectancy of this system a priori, and therefore technically I probably think that sample size is irrelevant.

I think time troublesome as a metric for sample size. Certainly time is important with respect to seeing a wide spectrum of market conditions to which a strategy must perform.

Rendering sample size as irrelevant to a statistical metric such as mathematical expectancy leaves many a strategy without any foundation.

 

[the hare]

What sample size would be sufficient?

I've already pointed this out to you at least 10 times.

The sample that you have is completely skewed with too many positive events and not enough black swan events. You probably need a hundred years of data, and even that wont be enough.

If you want to manage OPM taking a cut at a high watermark then your strategy is as good as it gets. If you want to fool idiots into paying for training, then its a great strategy.

If you want to get rich (long term) its a dumb way of going about it. I really do suggest reading Taleb if you are serious about this.

Thats the nature of black swan evens, they are unpredictable, bigger than you may imagine, and worse still, people seam to rationalise them after the event (in your case you rationalise by thinking if only you had more funds available etc)

Get on Amazon now and order Nassim's books, you'll be glad you did (even if you only use his arguments to improve your sales and marketing pitch)

 

[robster970]

Then read Tharp's book afterwards.

 

[scose-no-doubt]

Rendering sample size as irrelevant to a statistical metric such as mathematical expectancy leaves many a strategy without any foundation.

May I once again point you to the LTCM wiki page.

 

[Martinghoul]

What sample size would be sufficient?

This is actually a pretty deep question that I find myself pondering a lot...

The answer is (as always) that there is no answer and it really depends. Just like there's no such thing as a "risk-free rate" (contrary to conventional wisdom), there's no such thing as a universal horizon that validates a purely empirical approach (as an aside, this, IMHO, is one of the grand lessons of mathematics that we ignore). The only horizon that matters is your personal investment horizon.

EDIT: Just to make myself clear... I am saying that there isn't a horizon that is sufficiently long to validate an empirical trading strategy. However, one can definitely conclude that some sample sizes are much too small to make any conclusions worthwhile.

 

[Jack o'Clubs]

 

[amit1986]

The one I found that seems the simplest and most elegant is positive expectancy.

Yes, I suppose it would be okay to say a trading edge is a skill/strategy that has a positive expectancy.

Would it be satisfactory if I described my strategy in those terms?

I know how your strategy works - you've described it quite well. All I ask is for a few words on what unique skill or set of skills you believe you have that sets you apart from the crowd of other traders.