John Hill and backadjusting

Use this forum to discuss data providers like CSI, charting, or other non testing software.
Post Reply
kianti
Roundtable Knight
Roundtable Knight
Posts: 335
Joined: Fri May 02, 2003 6:10 am
Location: Florence - Italy

John Hill and backadjusting

Post by kianti » Tue Sep 28, 2004 12:10 pm

From another forum:
9/28/2004 5:59 AM I am wondering if someone of you guys use wealthlab to trade futures with a system. nearly every dataprovider provide backadjusted futuresdata. but this is not the realsworld. I read the book ultimate trading guide by john hill. he tested every system with the actual contract. he used the software excalibur. Now I am asking me, if it is possible to do it the same way with wealthlab? This means. Every time a contract expire, the system use the next contract and calculate the spread between the contracts in the p&l.
I have not read Hill's book and never back-tested a system using the actual contracts. I would like to hear any opinion and experience.

best regards,as ever

jankiraly
Roundtable Fellow
Roundtable Fellow
Posts: 52
Joined: Tue Feb 03, 2004 7:23 pm
Location: San Diego

Post by jankiraly » Tue Sep 28, 2004 12:37 pm

Maybe you are not aware that John Hill has discarded Excalibur and now uses Tradestation. He even writes books about how to use Tradestation:

http://www.amazon.com/exec/obidos/tg/de ... ingblox-20

I happen to own a copy of the Excalibur software (Hill used to sell it). Written in Fortran, runs on the Mac. Or should I say, ran on the Mac, before Y2K. It uses actual contracts and performs rollovers as actual trades. The rollover schedule was fixed, built into the software, and not user adjustable. And boy was it stupid; it didn't roll over anywhere near when I rolled over my positions in my account.

Anothe Excalibur afficianado, besides John Hill, was Randy Stuckey (vendor of Catscan, d/b/a "Mindfire Systems"). He used to publish his backtest reports using Excalibur. But he, too, has switched to Tradestation.

Tradestation does not and cannot use actual contract data and perform rollover trades. Sorry.

ksberg
Roundtable Knight
Roundtable Knight
Posts: 208
Joined: Fri Jan 23, 2004 1:39 am
Location: San Diego

Re: John Hill and backadjusting

Post by ksberg » Tue Sep 28, 2004 2:17 pm

kianti wrote:I have not read Hill's book and never back-tested a system using the actual contracts. I would like to hear any opinion and experience.
I have a custom proprietary backtesting engine that fits that description. It was designed from the ground up to test on actual contracts and vary rollover according to a number of conditions. For instance, it can roll to most popular contracts skipping less popular ones, or roll in advance of FND according to time and/or volume, etc. Roll policies were set on a per market basis (e.g. Crude rolls every month, while Eurodollar could roll on December-to-December contract). The goal was to simulate real trading decisions as close as possible.

I did find a big difference between actual and continuous contract, mainly in costs (which means that continuous contracts can introduce fantom compounding over actual trading). This type of impact can be simulated atop continuous contract by including roll factor costs, like VeriTrader does. This is the direction I've moved myself, simply because there are other complications from using individual contracts. I still get the ability to customize roll policies on a per-market basis.

One complication is correcting the discontinuity in long-range indicators. For instance, it's relatively easy to adjust a Channel Breakout system between contracts, it's much more difficult to adjust an 80-period moving average when the new contract may not have sufficient back data, or the price movement in the new contract doesn't accurately capture prior volatility. I've considered a hybrid use of continuous and actual contracts, but that still has the following problems ...

Another complication is dealing with vendor data errors. Face the facts, ALL VENDORS have some data errors. It is much more difficult to find and correct these errors when they are spread among thousands of individual contracts than it is to check a basket of 40 continuous contracts. As a result, backtesting on actual contracts are more susceptable to data errors.

I'd like to see more study around types of continuous contract and type of trading system. A system based on support/resistance prices probably requires different data than long-range indicators or even pattern studies. There are compromises made at every point in trading simulation. Trading data is an interesting study all in itself.

Cheers,

Kevin

kianti
Roundtable Knight
Roundtable Knight
Posts: 335
Joined: Fri May 02, 2003 6:10 am
Location: Florence - Italy

Post by kianti » Tue Sep 28, 2004 4:52 pm

jankiraly wrote:Maybe you are not aware that John Hill has discarded Excalibur and now uses Tradestation. He even writes books about how to use Tradestation
Thanks for the feedback
ksberg wrote: I have a custom proprietary backtesting engine….
I did find a big difference between actual and continuous contract, mainly in costs ….
One complication is correcting the discontinuity in long-range indicators….
Thanks for the feedback, after making adjustments in commissions and slippage for contracts with more frequent rollovers, I have the feeling that shortening the length of the breakout indicator increase the difference between actual and simulated results.

Best regards, as ever

SysTrader
Contributing Member
Contributing Member
Posts: 9
Joined: Mon Nov 24, 2003 4:15 pm

Individual Contracts

Post by SysTrader » Wed Sep 29, 2004 4:52 am

I am also using individual contracts (even tick by tick) for backtesting trading systems. Cleaning the data and removing "outliers" is really a time consuming and burdensome work. Some exchanges (like EUREX) have excellent tickdata. Using tickdata gives you an advantage in designing and testing short term trading systems. I have also tested LTTF systems on individual contracts (also using tickdata). Using trading rules with long lookbacks (eg 200days) requires the calculation of a ratio adjusted contract using the underlying individual contracts. So the signals are generated on the ratio adjusted data and the individual contracts are used for simulating fills and rollovers. The simulation results match better to reality. Slippage assumptions have the biggest impact on my simulations. I tend to be conservative with slippage.

kianti
Roundtable Knight
Roundtable Knight
Posts: 335
Joined: Fri May 02, 2003 6:10 am
Location: Florence - Italy

Re: Individual Contracts

Post by kianti » Wed Sep 29, 2004 10:36 am

SysTrader wrote:... on the ratio adjusted data ...
What's the difference between back-adjusted and ratio-adjusted futures contracts data?

best regards, as ever

vgs
Full Member
Full Member
Posts: 10
Joined: Thu Mar 25, 2004 8:42 am

Post by vgs » Sat Oct 02, 2004 6:40 am

What's the difference between back-adjusted and ratio-adjusted futures contracts data?
The following are quoted from CSI webpage:
You may choose to express the resulting series as a back-adjusted series, a forward-adjusted series or a proportionally adjusted series. The back-adjusted series preserves the current contract prices and adjusts all earlier series according the application of the roll trigger and the user's Representative Price choices. To prepare a forward-adjusted price series, the back-adjusted logic is engaged, but the cumulative delta effects are added back in, (after the fact), forcing the very first (earliest) contract to reflect actual values and leaving later contract series to be adjusted. This selection is sometimes used when back-adjusted series result in unwanted negative prices early in the distant past. The proportionately adjusted series, also known as ratio adjusted series, are actually back-adjusted prices which are adjusted, not by a cumulative delta quantity, but by a multiplicative approach which preserves positive prices into the past in a relational perspective.
Hope this one help. Best regards.

Kiwi
Roundtable Knight
Roundtable Knight
Posts: 513
Joined: Wed Apr 16, 2003 1:18 am
Location: Nowhere near

Post by Kiwi » Sat Oct 02, 2004 10:57 pm

The proportional approach (multiplicative) was promoted by Stridsmann in his book because it does two good things. 1st, as vegas pointed out, it prevents the data from going negative (look at crude oil with conventional back adjusting) and 2nd it keeps the proportions correct. What I mean by that is that if you take S&P contracts and back adjust back to 1987, although the absolute change in the index future is correct, its proportion of the indexes value at the time is highly distorted.

Personally, I prefer standard back adjusting (additive) because it does a couple of things relevant to traders. 1st it is close to what you actually do when you trade (if long you sell the old contract and buy the new contract at open or close (in a simple, lets not be too clever, case) so its just like backadjusting with open to open mapping. 2nd the price difference between two points stays the same as if you were trading it.

Most people end up staying with some form of additive adjustment, typically what CSI delivers straight out of the box.

Post Reply