Testing of Trading Strategies

>> Download Powerpoint Presentation <<

There are plenty of trading strategies which are commonly employed by Hedge Funds and professional traders that simply don't work in the long run. The goal of this website is to demonstrate that some widely propagated trading strategies don't work in the long run and add no value to potential investors.

My methodology to test the strategies is as follows: All trading strategies are simulated with a Matlab program that makes no exceptions and methodically applies a programmed strategy. In most cases I use as a benchmark a simple buy-and-hold strategy of the respective index, stocks or derivatives on which I apply the trading strategy. While it may happen that I bump into strategies that may have worked temporarily in the past, a strategy can only be deemed successful, if it works over a long period of time. Most importantly, the return it generates needs to be statistically significantly different to the benchmark on a confidence interval level of 5%, independent of the distribution of its returns. If this is not achieved, a strategy doesn’t work, no matter how popular it is or in how many ‘finance books’ it is described as a way to make money.
While I will continue to analyse new strategies in the coming year, there is no schedule of when and what I investigate. However, if you have any suggestions trading patterns and strategies that you find worth being investigated, don’t hesitate to contact me!

 

Methodology
To show that a strategy does not work I usually present the following two elements:


1. The simulation with the Matlab program generates a histogram, which shows by how much a strategy outperformed a simple buy and hold strategy when applied many times. The mean should be close to zero when the strategy doesn’t work. If the mean is above zero, we need to test it for statistical significance, which means nothing else than trying to find out the probability that the outperformance occurred through coincidence:


2. Statistical tests: To test for outperformance I apply non-parametric tests.

 

Feel free to comment or make suggestions anytime in scrolling down.

Nicolas Dickreuter







 

   
     

Comments

Name


Email (optional, will not appear in the comment)

Comment



 

   

 

 

Breattace

why not...

 

Breattace

Yes, really.

 

Zetaannoxow

All can be

 

Zetaannoxow

thanks for the interesting information

 

Dr Peter Randle

Hi Nicolas So glad to see that you have kept piano as an important interest. I will download a few during next week. I note you have the 3rd Movement of Beethoven's Moonlight Sonata listed which is how we met online many years ago - Compuserve in the classical music forum! Have a good week! Peter

 

Aloysius

Well done! Your simulations debunk a great myth, on which a large part of the finance industry is built. All trading systems are in fact complicated martingale variants, which, ultimately, can lead to financial collapse (as shown by the 2008 financial crisis)... Great music too!

 

Cinsacanush

In my opinion you are not right. I can prove it. Write to me in PM, we will communicate

 

Dario

Is it possible to add Nikkei index starting from begining of 1990? it would be great to see how bad is Japanese recession compared to this one Thanks

 

Gray

Your point of the policy makers slowing things down is right on. Take a chart of 1929 and match the Sept 3 high with the 2007 high. Then expand the timeframe in 2007 - 2009 by 7 to 1. One day in 1929 is 7 days today. I cannot explain it but I have been watching it since Oct 08 and it is very very very similar chart. If it keeps trending the Dec 9 high in 29 is equivalent to a high coming in late Aug/early Sept. The price is about 35.5 to 1.

 

Jackie Ann Patterson

Dear Nicolas, Thanks for the delightful music! I downloaded several months ago. iTunes just shuffled up your Chopin Etude, which is my favorite of this collection. Reminded me to visit your site and say thanks! Jackie

 

Pharma270

Very nice site!

 

Grispino

Good morning. Nice work.,

 

Satherland

Man, that ain't funny. You did me right by letting everyone know in your blog I was simply waiting in line for. I'm cool on the number. You two keeping in touch??,

 

Leaviston

Just stopping by to say Hello!,

 

Khaysavang

Hummina Hummina huh! Keep up the good work!!,

 

Tim Mallon

Peter is correct, there are of course divergences, but the internal dynamics will drive prices that will unfold into very simular price patterns. Independant trades made by millions of people still follow power laws that govern fractal geometry and symmetrical out comes. Simple social cues were everywhere for anyone to see. Each generation becomes a little more myopic untill the cues are dismissed as, (its different this time) then and only then are extreem tipping points reached ,we have arrived and you aint seen nothing yet.

 

Aabakken

You do some amazing work. Wishing you much continued success and hope to see you out at an event in the near future.

 

Aabak

amazing! :)

 

Aabach

your work is amazing.

 

Peter Lockhart

These are both Post Bubble economies. You could also compare the Nikkei from 1989 with the Dow and they look identical. Post Bubble Economies are after asset Bubbles rupture and the deflation and deleveraging then leads to Depression. It happens every time and its happening now.

 

Razor

Charts might look a lot more similar if hinged on August 11, 2008 as day 0 rather than October 09, 2007 - i.e. focused on EW 3 portion of decline rather than preceeding EW 1 and 2. Could you perhaps post this as a sensitivity?

 

MTRIG.com

I have to say I found this post very intriguing. I have back tested many strategies (sell in may, best months, but not day to day). Very good! www.MTRIG.com

 

Leonard

Thank you! REally cool. ON my default page. :-)

 

Eric W

This comparison is the only match; it won't fit with any other recession. If you look also at the societal conditions and factors, we are in the Spring of 1930. Massive debt unwinding, monetization of the debt, ARM loan resets, unemployment, etc. Also, we now have half the population in the market via 401K's and IRA's. If there is another leg down below the March lows, risk appetite for the stock market will be zilch and hundreds of billions (trillions) will leave the stock market and not come back, pushing it lower still.

 

Eric W

This comparison is the only match; it won't fit with any other recession. If you look also at the societal conditions and factors, we are in the Spring of 1930. Massive debt unwinding, monetization of the debt, ARM loan resets, unemployment, etc. Also, we now have half the population in the market via 401K's and IRA's. If there is another leg down below the March lows, risk appetite for the stock market will be zilch and hundreds of billions (trillions) will leave the stock market and not come back, pushing it lower still.

 

Doug Ewald

I've been following this graph, why do you show such a drop today? Market went up 6-1-2009

 

Question:

June 1, 2009 What gives? The 2007-today graph just changed radically from what it looked like yesterday. Is it broken, or did it never work and you are being goofy?

 

Miraenda

Those who think this chart supports anything specifically, you might want to read the page source. It has for the content: "Shows how easily we can be fooled in believing that there is a trend in a stock price" The purpose of the chart is to say there is no purpose from what I gather.

 

gem

I was trying to, like everyone else, get something for nothing and go on your auction site which is hammered so I've given up. Found this one instead. Very interesting person... Very interesting website.... Plus I've downloaded your fantastic playing so I guess I did get something for nothing! Good luck with your future business venture. x

 

Lou La Torre

Im seeking an invester/publisher to publish my book based on my personal experience. Have a look at my website www.justiceforcanadians.ca The book will name the Canadian Federal Conservative Politician who was my lawyer, it will name the now retired Judge who threatened my life, the so called legal binding document with a FRAUDULENT SIGNATURE which my lawyer claimed that I signed, the court house file that was raided and half the documents missing especialy the documents dated September 23, 1999. The court house clerk who was kind enough to give me a hand written note and verify that I was at the court house waiting for the documents while she searched the file (What was left of it, and now missing.) The Attorney General of Ontario Canada who stated in their letter dated February 2004 "We will not be calling a public inquiry into the auto industry at this time." Which gave the perception they will do so, some time in the near future, then they realized it was the Federal Conservative MP partner who represented me at the time the Federal Conservative MP was the Attorney General Of Ontario at the time all this was ocurring. On top of that the Law Society of Upper Canada stated in their analysis "The lawyer did nothing wrong." People of the world think Canada is not a corrupt country, should think again. Let me get my book published and the world will soon realize Canada has it's problems too. Cheers, Lou P.S. Like the composition, kudos!

 

Doug

Haha, can't believe people are actually asking you for money??!! Beggers... Good job with the auction site, its crashed today as there are so many people on it! Your details - Phone number, address can be found through your domain, best take that off or you'll be getting people coming round your house!!!

 

amy

Hi, how are you? this will sound very cheeky so please excuse me, but i read that your giving away a fourtune, and was wondering if you could help me. I recently went out to L.A to complete my trainning as a make up artist, to do this i had to take out a loan of £15,000, however the school was so bad and they let me down on accomendation, so i had to try and find somewhere inbetween classes, it was so hard, but i always say if the harder you work the better the rewards, so i continued, but i had to leave the course two weeks b4 i was due to finish as i didnt have enough money to complete my tuition. I was heart broken, even though i didnt think much of the course, i was told it was the best and i was looking foreward tp working out there, anyway after doing some research out there i found that there was a school that is world renowned, and was wondering if you would be intrested in paying my tuition?, i know i told you that this would sound cheeky, but i am saving for it and working 19hrs a day, but its going to take a couple of years, and to be honest i can't wait, this is my dream, all that i have ever wanted to do, and i can't take out another loan. I would pay you back, i will hand over all i own to you untill i do, and i know that i am good at this, to sound really sad, its all i know!, anyway, i am sorry to bother you, but you are my last hope. i h ope to hear from you soon, best wishes amy x

 

martin brown

Feel free to give me some of you hard earned cash please. I'm in debt and struggling, get peice of mind from helping me out. x

 

Armando Alizo

I'm not sure I agree with this! I ran a back test with the $SPX and a simple 200-day MA (Buy at next day's close if the index crosses above the MA, sell if it crosses below). I started in 1942 as that is the earliest date I have and I used the S&P 500 Tot Return index, but I got similar relative results using the unadjusted index. =========================================== BUY & HOLD CAR: +11.05% Max Drawdown: -55.20% Ulcer Index: 11.78% ========================================== 200-DAY MA CAR: +12.19% Max Drawdown: -21.07% Ulcer Index: 6.09% ========================================== So it certainly does seem that using a MA is a good strategy. If nothing else is reduces Max Drawdown!!

 

Scott

Re the x-axis...I think we are about 400 market days into this...rather than calendar days

 

Armando Alizo

Dick, While I agree that you can generate charts that "look" like trending securities using random variables, that doesn't prove that Trend Following as such is an invalid strategy. There are many ways to define a trend - some may be profitable in some markets and not in others. Naturally ANY approach (trend based or not) must be properly formulated and back tested. I do agree completely with your comments regarding TV "gurus" that use charts as if they were tea leaves to predict the future. The future is unpredictable, and in my view it is actually COUNTERPRODUCTIVE to attempt to make forecasts. Such approaches are utterly subjective, inconsistent, and unprofitable.

 

Me and just Me

I will print this out on soft paper and wipe with it. I like my monitor enough not to wipe with it. Is there a point to this graph? I would like to see a video of how this was put together. There will still be idiots out there that will look at this a get scared. CNBC might actually publish your work. I am an idiot for commenting on this. Have a great dumb day! Minyanville who linked to this graph and people like you should have links on the National Enquirer website.

 

Nicolas Dickreuter

Andy, you only need to count the trading days and not the calendar days, that is why 400 days is correct.

 

Michael A. Weinstein

As the saying goes, "Those who cannot remember the past are condemned to repeat it". This is not to say that we will, (or must), travel down the same path as in the 1920’s. We now have better and faster communications, (we no longer have to depend on the ticker), computer trading, including much more sophisticated analysis tools and stop-loss controls. I believe the 1929/2009 analogy to be nothing more than the human predisposition to insist in connecting two dots on a page with a straight line. maw

 

Michael A. Weinstein

As the saying goes, "Those who cannot remember the past are condemned to repeat it". This is not to say that we will, (or must), travel down the same path as in the 1920’s. We now have better and faster communications, (we no longer have to depend on the ticker), computer trading, including much more sophisticated analysis tools and stop-loss controls. I believe the 1929/2009 analogy to be nothing more than the human predisposition to insist in connecting two dots on a page with a straight line. maw

 

andy harris

The x-axis scale seems to be off: if the 100% value was market peak (14,093 on or around 10/8/07, I believe), then we should be closing in on day 550--yet the graph appears to chart us at around day 400. Am I reading this chart wrong? Or is there a miscalibration? I would certainly appreciate any additional explanation. Cheers, Andy

 

vdavisson

Good comment Randy

 

G D

You have the Dow going down for a long time now but it has bounced 25% up in the last month?

 

soupreme

Very interesting. I've always wanted to see an analysis of this considering how much weight is given to certain days of the week by traders and expected market behavior for them (i.e. Turnaround Tuesday, Black Friday, etc.). Obviously if enough trade with that thought there has to be some influence there, and there does seem to be some significance in the occurrence of crashes/corrections in the month of Oct.

 

Orvinfive

you should have started the chart in March 2000, not in 2007, in my opinion....

 

Randy Gearhart

Murray Rothbard's book "Great Depression" identifies that during the summer of 1931 the general population looked upon all the actions taken by Hoover's government to prop up crop prices, farm credit, and wages with approval and believed that it would be sufficient to stave off further decline. Laws had been passed that would allow the fairly new Federal Reserve (born in 1913?) to use treasuries to prop up the balance sheet so that capital injections to the banks could continue. This period of 1931 would roughly correspond to the period we are entering now. The market advanced steadily during the early part of that summer. I'm not sure, but I don't think that Bernanke has ever read Rothbard's book. The book "Great Depression" that Bernanke refers to is focused on the FDR era; not the era and activities which led to the worst part of the 1930s crisis.

 

Anna Niederhaeuser

Bravo! Warst ja immer das "Wunderkind" in unserem Grueppli... Natuerlich gebuehrt auch dem lieben Dori ein grosses Lob.

 

Paul Piper

I am interested in your premium content. May I see it?

 

Vincent Bourgeois

Dear Nicolas, Very interesting website... Somehow I have not been able to listen to the above, I will try from home. Appologies about my Chopin (or Pollini) comments, probably a little misplaced. Regarding trading strategies, have you had a look at buying / selling stocks with an increase / decrease in the PAY OUT ratio of their dividend, on the basis that this gives forward looking information and that management is the best one to judge about the business ?

 

BingeLili

Interesting place :) Like it.

 

balu

dear sir, I am from doing,i am doing trading in commodity market , i am new to trade,i see through all details from your website,in trading the software were simulation,if you have any idea for doing sucess in trading please give me suggetion,its worthfull for us or any other method for sucessful for trading thanks & raegrds balu

 

Miss Finance and Equity

Well done, my Mr. favourite Student :-) You exceeded my highest expectation in turning the whole paradigma of investment strategy up side down. I'll give you the highest mark and the Equity Premium for such an enlightening ideas!

 

Mike

Awesome! Lovin' the Chopin Etude...

 

Ananymous

That's kind of amazing! Wondering what all those analysts are doing the whole day!

 

Laurent

Interesting view. But that's kind of surprising I think!

 

Mike

Good job! Like your site!

 

Anonymous

I must confess that I'm slightly baffled: what precisely is the connection between "playing the piano" and "trading strategies"?