Forums  > Books & Papers  > Quantpedia - the encyclopedia of quantitative trading strategies  
     
Page 1 of 1
Display using:  

radvan


Total Posts: 4
Joined: May 2015
 
Posted: 2015-05-05 15:20
Hello, I think this is an appropriate place to introduce ourselves as I hope our company Quantpedia.com can be useful for you.

We are quantitative trading research startup with a goal of continually building database of quantitative trading strategies derived out of the academic research papers. We read a lot of papers (from research portals, financial journals, universities etc.), select the best and extract trading rules in plain language, performance and risk characteristics and various descriptive characteristics (the instruments used for trading, markets traded, backtest period length etc.). Users can screen categorized strategies, examine related strategies or review visualized comparisons.

A lot of the strategies (and related papers) are readable for free, some of the higher performing or more unique strategies are accessible after subscription.

Let me know it you have any suggestion for improvement ...

Tradenator


Total Posts: 1570
Joined: Sep 2006
 
Posted: 2015-05-05 15:29
You are about a decade too late to be the first.

ThomasJ02


Total Posts: 33
Joined: Feb 2009
 
Posted: 2015-05-05 17:20
What are some alternative / competitor sites?

pj


Total Posts: 3305
Joined: Jun 2004
 
Posted: 2015-05-05 17:34
Google

OFFENDERS WILL BE TERMINATED

deeds


Total Posts: 341
Joined: Dec 2008
 
Posted: 2015-05-05 17:49

Such a site seems helpful as a shelf life indicator - in that it exhibits which strategies have been highlighted and exposed to what appear to be some investors with depth, and it also gives some sense of how long.

goldorak


Total Posts: 979
Joined: Nov 2004
 
Posted: 2015-05-05 19:28
This website has been up for a quite a while now. A very good tool. While people waste their time digging ideas from it they do not compete for real in the markets.

If you are not living on the edge you are taking up too much space.

finanzmaster


Total Posts: 96
Joined: Feb 2011
 
Posted: 2015-05-07 13:06
The idea seems to be very interesting, but:

"How we do it"
>We use a great number of finance research resources all over the
>world including research portals, financial journals, universities and
>conferences.
...
>By the end of each year thousands of financial papers have been
>investigated in this way.

I never considered myself a slowcoach but for me it usually takes several days to understand a paper in detail (not just to grasp the main idea).
How much human power do you have to "use a great number of finance research"?!

And how about the fact that "who knows, doesn't publish and who publishes, doesn't know".
E.g. a Professor with a good head on his shoulders: He said he's still curious about whether technical analysis really can beat the market after adjusting for risk--a question that his latest paper doesn't attempt to answer. If he and his colleagues discover that it can, he says, "We might never publish that paper."

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

goldorak


Total Posts: 979
Joined: Nov 2004
 
Posted: 2015-05-07 17:32
Because papers like most of the ones exhibited can be read in less than 30'...

If you are not living on the edge you are taking up too much space.

radvan


Total Posts: 4
Joined: May 2015
 
Posted: 2015-05-08 15:35
We are 3-4 researchers who are reading and analyzing papers. You are able to sift through academic research faster then you think, most of the papers aren't interesting enough to be published so you spend 90% of your time reviewing the best.

I do not agree wih the quote : "who knows, doesn't publish and who publishes, doesn't know". That's common myth on internet which is not based on truth. In reality a lot of good strategies are published and a lot of hedge funds are very transparent. The reason why they do it is because clients want it after 2008 and after Madoff. And hedge funds realized that it is perfect advertisement. You show what you know, how you do it,you will show your expertise and you will get a lot more assets under management.

Also, interesting papers are sometimes written at universities. You have better chance to be hired by top class fund if you have good publication in top class journal. So it makes sense for Phds to publish really interesting things.

I do not say that ALL great ideas are published. I say that sometimes a great ideas are published and we are trying to find them.

radvan


Total Posts: 4
Joined: May 2015
 
Posted: 2015-05-08 15:42
"Because papers like most of the ones exhibited can be read in less than 30'..."

Only the most common strategies are published for free. These are common knowledge. The most interesting papers are in premium section. Users can get an overview of database by using "Charts" section:

http://quantpedia.com/Chart
http://quantpedia.com/Chart/Complexity
http://quantpedia.com/Chart/Performance

goldorak


Total Posts: 979
Joined: Nov 2004
 
Posted: 2015-05-08 18:44
Could you please post a link to one article dedicated to a trading strategy that takes more than 30' to read?

And yes, I admit it, I always skip the "Existing literature/prior works" section in articles written by economists.

I do not like your charts on complexity. From simple you move up to complicated, not to complex. Complexity is a different thing.



If you are not living on the edge you are taking up too much space.

Tradenator


Total Posts: 1570
Joined: Sep 2006
 
Posted: 2015-05-08 19:05
So have you made some sort of financial donation or other contribution to NuclearPhynance to justify your advertising here? Do you have clearance from SIP?

Tradenator


Total Posts: 1570
Joined: Sep 2006
 
Posted: 2015-05-08 19:13
What kind of work have you done on the secular transition from falling rates (eg 30 years of Reaganomics) to rising rates (no catchy name yet)? I'm interested in what to expect for the next 20 - 30 years in terms of the effect of rising rates on commodity carry (convenience yield), equity valuations, bond volatility, etc, etc.

EDIT: Do you think about what might be topical for readers or is that irrelevant?

radvan


Total Posts: 4
Joined: May 2015
 
Posted: 2015-05-09 15:26
"Could you please post a link to one article dedicated to a trading strategy that takes more than 30' to read?"

I like for example:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2378969
http://herkules.oulu.fi/thesis/nbnfioulu-201211141056.pdf
http://www.be.wvu.edu/div/econ/work/pdf_files/09-14.pdf

I personally prefer simple ideas not complicated one. Our experience shows that simple ideas have better probability to be profitable out of sample then some systems with complicated fancy math. What matters is underlying fundamental idea which can be usually captured by simple rules.

"What kind of work have you done on the secular transition from falling rates (eg 30 years of Reaganomics) to rising rates?"

I would call this "scenario analysis". We do not do scenarios so I am afraid I do not have anything meaningful to say about this subject.

goldorak


Total Posts: 979
Joined: Nov 2004
 
Posted: 2015-05-10 19:32
I see the good old "out of sample" guard.

If you are not living on the edge you are taking up too much space.

goldorak


Total Posts: 979
Joined: Nov 2004
 
Posted: 2015-05-10 19:49
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2378969

I decided long ago not to have any contact with options anymore. So I won't comment.

http://herkules.oulu.fi/thesis/nbnfioulu-201211141056.pdf

fig 21, p.53 was found in less than 30 seconds. It says it all. Vertical classification.

I personally love the paragraph few lines below the figure: Putting these results into practice would be pretty straight forward. One would need a futures account, and up to date futures contract prices and open interest information.


Hmmm they probably forgot to mention a sizable wealth too. You know, the one you are going to turn into a negligible wealth using our fantastic article.

http://www.be.wvu.edu/div/econ/work/pdf_files/09-14.pdf

30 minutes to read this? Is this a joke? Am I the only one thinking this is just a pale iteration on the "good" old momentum / mean reversal strategies, but applied to a selected few currencies?


If you are not living on the edge you are taking up too much space.

goldorak


Total Posts: 979
Joined: Nov 2004
 
Posted: 2015-05-10 19:50
> What matters is underlying fundamental idea which can be usually captured by simple rules.

I would buy anything capturing a fundamental idea, especially if strategy/product's name contains words like "enhanced" or "plus".

If you are not living on the edge you are taking up too much space.

TakeItAndRun


Total Posts: 88
Joined: Apr 2010
 
Posted: 2015-11-28 12:19
http://herkules.oulu.fi/thesis/nbnfioulu-201211141056.pdf
"Pairs Trading The Commodity Futures Curve"
I cannot reproduce the results of the Master’s thesis of Antti Nikkanen (supervisor Hannu Kahra).
He claims to get an annualized return of above 6% over more than 15 years of backtest with a Sharpe ratio of 3.

Here is the idea of his strategy:
For the 20 most liquid commodity futures, in case of backwardation, sell spread; otherwise buy spread (for clarity “buy spread” means buy back month future and sell front or near front future). Hold one month. Close position. Iterate.

I think the real Sharpe ratio is close to 0 from 2000 to 2015. I cannot imagine getting a Sharpe ratio of 3 with only the present term structure as information.

My guess is that Antti got this kind of results using the future term structure!


elvolkan


Total Posts: 1
Joined: Jun 2016
 
Posted: 2016-06-01 17:16
I have tried and failed too.

The promise seems fair, the method looks simple, but Mr. Nikkanen seems to be peeking into the future, or else giving misleading rules.

http://jultika.oulu.fi/files/nbnfioulu-201211141056.pdf

I started replicating NG, then CL, then all 20 commodities, just to get a 0,8 SR.

Any opinions on this?

goldorak


Total Posts: 979
Joined: Nov 2004
 
Posted: 2016-06-01 17:58
Welcome to the real world. The peer review system has been a successful history for centuries. However when it comes to economics, finance, medicine, psychology and other pseudo-sciences, unfortunately the peer review system has miserably failed. The only thing of importance for any peer reviewer is to make sure you and your friends get a citation from the article. Publish, get citations or perish. Not entirely their fault though.

Studies not reproducible, unavailability of data, no open code published, use of future information (eg there is a positive correlation between X and Y, let's see if a strategy using X to predict Y has good results), everybody can observe and trade an asset at the same time, overuse of p-values, data snooping, etc...

I think I could have spent my entire life publishing counter studies of all the BS published over the last 30 years, and JoF is probably the worst of all.




If you are not living on the edge you are taking up too much space.
Previous Thread :: Next Thread 
Page 1 of 1