Forums  > Trading  > A case study to show by means of statistical test this guy (it is NOT me) can beat the market and that it is not an accident  
     
Page 1 of 1
Display using:  

finanzmaster


Total Posts: 125
Joined: Feb 2011
 
Posted: 2016-10-09 17:27
I often hear (even from financial professionals) that a track record may be a pure luck and in either case one needs at least 10 years of track record to "boast" with it. I show that (given a sufficient number of trades and independence of the market regime) 2 or 3 years are enough to make a track record proven (or at least statistically significant)

Though the article is primarily intended for non-professionals, I think it will be interesting to professionals, too ... at least I experienced a lack of ability to judge a track record even among fund managers.

Einstein – a star trader on Wikifolio, who can beat the market

www.yetanotherquant.com - Knowledge rather than Hope: A Book for Retail Investors and Mathematical Finance Students

Rookie_Quant


Total Posts: 741
Joined: Jun 2004
 
Posted: 2016-10-09 17:55
what is the appropriate benchmark and are you controlling for the actual risks taken in the strategy? Looking at the DAX but talking about a micro-cap strategy is off-putting at the outset.

I skimmed the report but that's my first comment.

Where is the alpha analysis? And dont just regress "Einstein" on the DAX and show me a positive intercept.

Perhaps the ex ante probability of trades is 50/50, although that in and of itself is meaningless if the relative magnitude of wins and losses is asymmetric. Heads I win $1, tails I lose $5....look at me though, I flip heads 75% of the time!

Also, the icon at the beginning looks like a tri-testicle. Just sayin'

"These metaphors and similes aint similar to them, not at all." -Eminem

Azx


Total Posts: 26
Joined: Sep 2009
 
Posted: 2016-10-09 17:56
Well, there is a very reasonable explanation why the results are significant, If they weren't you wouldn't have picked them to do your test. Also, the binomial distribution is not really the proper way to analyze trades. It is pretty easy to exchange a high win rate with average win/loss and vice versa.

Furthermore, there is still some debate about how much can actually tell us about .

finanzmaster


Total Posts: 125
Joined: Feb 2011
 
Posted: 2016-10-09 20:19
@Rookie_Quant
>what is the appropriate benchmark
The "genuine" question sounds a little bit different, namely, whether a benchmarking is an appropriate approach :)
Einstein obviously optimized the absolute return, and doesn't care about benchmarks.

> Looking at the DAX but talking about a micro-cap strategy is off-putting
Yes, the DAX is not a perfect proxy for small/micro-caps but it is still an acceptable proxy for the German economy. I refer to DAX not as to a benchmark but as indicator of market regime (bool, bear, flat).


> Heads I win $1, tails I lose $5....look at me though, I flip heads 75% of the time!
Correct! but Figures 2 und 2a show that this is not the case.

@Azx >It is pretty easy to exchange a high win rate with average win/loss and vice versa.
Yes. But see Figures 2 und 2a :)





www.yetanotherquant.com - Knowledge rather than Hope: A Book for Retail Investors and Mathematical Finance Students

LeDude


Total Posts: 1
Joined: Oct 2016
 
Posted: 2016-10-24 13:03
I noticed that the article speaks about 1916, of which 883 are sells. Then those 503 of those 883 apparently were positive.

Well, if you just look at the sells you'd have a directional tilt depending on where the market went that period.

And how is each "hit" counted - if I have a position where I made money, can I sell that off in many increments (while losing positions are closed in on trade) and therefore get a great hit ratio?

finanzmaster


Total Posts: 125
Joined: Feb 2011
 
Posted: 2016-10-25 21:44
>Well, if you just look at the sells you'd have a directional tilt
>depending on where the market went that period.
As a matter of fact Wikifolio doesn't allow shortselling ... well, there are several ETFs/ETCs for going short, but you still need to by them. That's why considering SELLs only does make sense.
And there is no dependence on whether the market (DAX) grew or fall, Einsteins portfolio has always been growing (compare Figure 1 with Figure 3)

>can I sell that off in many increments (while losing positions are closed in on trade)
Then your profitable trades would have much less weight in portfolio context than the lossy trades. Figure2a tells us that it is not the case.

www.yetanotherquant.com - Knowledge rather than Hope: A Book for Retail Investors and Mathematical Finance Students

rftx713


Total Posts: 95
Joined: May 2016
 
Posted: 2016-10-26 03:22
...

finanzmaster


Total Posts: 125
Joined: Feb 2011
 
Posted: 2017-09-23 14:52
Another star trader on wikifolio, from my point of view even better than Einstein on the risk-adjusted basis.

https://letyourmoneygrow.com/2017/09/23/hbecker-another-star-trader-wikifolio/

BTW, Einstein kept (as expected) his pace and approaches the return of 2500%!

www.yetanotherquant.com - Knowledge rather than Hope: A Book for Retail Investors and Mathematical Finance Students
Previous Thread :: Next Thread 
Page 1 of 1