Q: VT talks about having several well researched, non correlated trading systems. I have been trying to figure out what instruments are not correlated with each other. So far, those I've thought of appear to be inversely correlated with each other, or in some way related to each other.

For example, when USD goes down, Gold goes up. When oil goes up, stocks go down. So if I have systems that trade both USD and gold, it seems that what I have are 2 systems that are inversely correlated to each other.

Could you give me an example of systems which are non correlated to each other?

A: You are confusing trading systems with markets. They are not the same. However, you could have a long term trend following strategy that trades oil and stocks. Your strategy (which is what I'm talking about) would be the
same in both markets, the results would be different. What you are looking for is non-correlated strategies.

Even market types (up-volatile vs down quiet) are not the same as markets. Thus if you find a system that works in a volatile quiet market (i.e., some option strategies such as a calendar spread), it will not be correlated with one that works well in a long term bull (volatile or quiet) environment (i.e., my efficiency strategy). Similarly, there are swing trading strategies that work in most environments that have little relationship to long term buy strategies or long term shorting strategies.

You can even think of the market as a sine wave. Some strategies work well when the trend is underway. Other strategies try to get onto the trend when it first starts by picking bottoms, other strategies pick tops and go short, while still other strategies go short when a down trend is well underway. And these sine waves to describe the market can be long term sine waves or very short term (i.e., intraday).

Posted by Van K Tharp at 05:09 PM | Permalink | Comments (0) | TrackBacks (0)

Q: 1. What is the best way to investigate what the correlation between two trading systems is? By comparing both equity lines? Or by comparing the day by day trading results of both systems? Or by comparing the algorithms? Or maybe something else?

2. What do you think is the best way to investigate what the correlation between two instruments is (for example DJIA and SP500)? The historical quotes? Or by comparing the day by day price changes? Or maybe something else?

3. And finally, what do you think is the best choice: Diversification by using ONE trading system for TWO uncorrelated markets or using TWO uncorrelated trading systems for ONE market?

A: All your questions are based upon some assumption of what best means? And I have no idea what it means to you… only to me. The following comments should indicate why.

When I suggest that people use non-correlated trading systems, it really means that there is very little relationship between the concepts behind it. Thus, a trend following systems that looks for a well established pattern before it entered, should have very little correlation with a bottom picking trading system that looks for extreme down moves and then attempts to catch sharp 1-3 day reversals.

Now both of those systems could make money every week and would appear correlated, but I would say that they are not because the concept you are trading is so different. That’s really my definition of non-correlated systems and my criterion. But it might be quite different from yours. As far as equity curves are concerned, I’d like all my systems to be highly correlated in that they ALL have a LARGE PROFIT every month.

Why would you want to measure the correlation between the S*P 500 and the DOW. The S*P 500 contains generally contains the DOW, does it not?

Again, for question 3, I have no idea what you mean by best. What are you trying to accomplish through diversification? That might define what’s best. And why would you want to diversify when some of the best investors (like Buffett) say that diversification is a substitute for ignorance.

Posted by Van K Tharp at 03:40 PM | Permalink | Comments (0) | TrackBacks (0)

Q: I am half way through the fifth volume of the Peak Performance Course.

My question is about tampering with a trading system. The course content seems to suggest that the top traders follow their trading system verbatim and through superior money management and risk control, they are able to achieve consistent profits. But, my trading system supplier is telling me to use discretion when using their trading system. For example, to stand aside during the FOMC meeting week, to avoid trading during large range days. To me, this sounds like tampering with the trading system, which a top trader would never do.

But I am thinking, maybe tampering is acceptable if you use additional filters ONLY TO OVERRIDE trade signals and NEVER TO GENERATE them. That is, only accept signals that are generated by the trading system, but override them if the additional filters suggest that you do so.

How does that sound?

A: No trading system works in all market types and most are discretionary. You should know what the market type is (see the latest update in Tharp’s Thoughts) and how your system performs under those condition. And not trade it if the conditions are poor.

I usually recommend in the Peak Performance Course, for example, that you not be discretionary until you know what you are doing.

Posted by Van K Tharp at 07:52 PM | Permalink | Comments (0) | TrackBacks (0)

Q: The formula for the 'Average Directional Movement Index' (ADX) on page 268 (second edition of Trade Your Way), is this correct? I have been unable to reconcile my own calculations with those found at various financial websites. The majority of ADX calculations I have found involve exponential moving averages but 'Trade your way...' states that only a simple moving average is required?

A: The simple moving average is fine. If you want to go to the source on ADX, go to Wells Wilder's book, New Concepts in Technical Trading. Actually when you think of it, if you really want the average true range (meaning a measurement of volatility over say the last 14 days), why would you do an exponential moving average? That would not give you an accurate measurement of what's happening in terms of volatility.

Posted by Van K Tharp at 05:41 PM | Permalink | Comments (0) | TrackBacks (0)

Q: I have been studying the commodities markets for mostly my own edification for over 8 years with a brief attempt at trading a few contracts three years ago before realizing I needed more formal training if I wanted to trade successfully. To make a long story short, after reading approx. 20 books, purchasing a software program (“Mechanica”), end-of-day data feed and taking one on-line course prior to yours, I am getting ready to start testing a system of mine with a paper trading account through an on-line broker. My question – I have saved up 20k with which to trade. I tend to gravitate toward short term trend following, mechanical systems. From what I can gather from the recent course, most trend following strategies are not very compatible with an account less than 100k. I am not opposed to day trading strategies (i.e. swing trading, etc) but I currently work all day and need to make my decisions early in the morning (I live on the west coast) or evening. Do you have any recommendations for me with regards to types of systems I can test that might be more compatible with my current living situation?

A: You need to think about your risk and at first make sure that your risk is no more than 1% per position, perhaps ½% when you start trading. There are some futures contracts that you can probably trade with $20,000 but it depends upon where your system says to put your stops.

Let’s say you trade a corn contract and that corn is currently trading at $6.00. Your 5000 bushels of corn will cost you $30K which you can easily do on margin with a $20K account. But where is your stop. Let’s say it is 10 cents away at $5.90. Your risk is 10 cents time 5000 bushels or $500.00. You’d be risking 2.5% on your one corn contract. Thus, you can’t even trade corn unless you want to use a 2-3 cent stop and your system probably isn’t that tight.

So what you are looking at is finding instruments that you can trade where you risk is no more than $200 per position. You could buy a $20 stock with a stop $5 away. That would give you $5 risk per share. And if don’t want more than $200 in total risk, then you could simply purchase 40 shares. That’s how you would decide what you can trade and your position sizing.

1) Determine what you system says your stop would be.
2) Knowing that your risk should not be more than 1% of your account or $200, divide your risk per unit into 200 and see what you position sizing is.
3) If you can do that amount, then you can trade that instrument.

Posted by Van K Tharp at 07:43 PM | Permalink | Comments (0) | TrackBacks (0)

Q: I have read and re-read (as you suggest) your book "Trade your way to financial freedom". It think it is great material, it is such a mind-opener.

However, I would like to clarify one point that it is yet not clear enough to me. It is about the Percent Volatility Method. If I use such Position Sizing Technique, then the Stop Loss is set by this position sizing method, I mean, the strategy must have a Stop Loss that is equal, for instance, to the 20-day SMA of ATR. If that is the stop loss, then the model is the same as the % of my equity.

I am getting confused. Let;s say that, prior to Position Sizing, my strategy has a Stop Loss that is 20-day SMA of ATR. Thus, the % of my equity and % volatility method are just the same. And if my strategy has no stop loss (let´ts say), if I use the Percent Volatility Method, it will be the method that will calculate the Stop Loss and place it 1ATR below the entry price.

A: Position sizing and the initial stop are two entirely different issues. The stop is NEVER set by the position sizing.
However, if the stop is 1 ATR, then percent risk and percent volatility are the same.

Posted by Van K Tharp at 04:53 PM | Permalink | Comments (0) | TrackBacks (0)

Q: Safe Strategies talks about bear mutual funds but recommends not to use the leveraged funds - why should one not use the leveraged funds and presumably can one use ETFs instead of a bear mutual fund?

A: That strategy specifically controlled the maximum exposure that you would have and the amount of risk you would have. If you used a leverage fund, then all the metrics no longer work and the risk is quite large.

Posted by Van K Tharp at 04:46 PM | Permalink | Comments (0) | TrackBacks (0)

Q: I'm a professional money manager running an open-ended mutual fund, with subscriptions and redemptions on a daily basis, which has a mandate to be at least 95% invested in stocks at all times.

How do you think about position sizing in a case like mine (which can obviously be applied to anyone with a portfolio whose assets are subject to inflows/outflows)? If I start off with USD 10M and invest it equally into say 50 stocks with a 2% position in each, what happens if after say 3 months the portfolio receives a USD 5M inflow. During those times the stocks have all moved, and some are trading below my 'in' price, others have done well. My feeling has always been that rather than add 50% more to each and every position, you add only to the 'winners'. The problem is that you then end up with a portfolio which is no longer equally-weighted, as you've maybe doubled the investment in say 25 winners and done nothing with the say 25 losers.

Sometimes I feel that unless you have a discreet pot of money where there are no inflows/outflows, the system breaks down because of things like this. What do you suggest?

A: You are in a big bind without your mandate to be 95% invested at all times and my guess is that many money managers with such a mandate will be out of business within the next ten years. What are you going to do when the baby boomers retire in a big way and you constantly have a net outflow of funds? And when that happens to all the fund managers, the major indices (that everyone is holding) will go down big time.

My big question to you is how can you get your mandate changed? It’s something you’ll have to do eventually and it's much better now than later.

Anyway, I have a specific solution for you in my new book The Definitive Guide to Position Sizing. Basically you have to buy the index that is your benchmark. Position sizing will be how much you under or overweight any stocks in the benchmark, with an underweight being a short position. Specific details are in the Definitive Guide (out soon) and it’s also in an article I did with a mutual fund manager (An Interview with Steven O’Keefe) which appeared in Market Mastery and back issues are currently available for sale through IITM.

Posted by Van K Tharp at 07:28 PM | Permalink | Comments (0) | TrackBacks (0)

Q: My question is about conflicting goals.
Goal number one is to cut losses and let profits run.
Goal two is make a certain dollar profit per week.

My conflict is that I have 2 profitable trades on. If I close them both, I will achieve my specific dollar profit for the week on Wednesday. But I want to let my profits run as well. One trade is weakening and I wouldn't put the trade on today using my entry signals. The second trade is still showing good entry signals and I would put the trade on today.

What I finally decided to do was close both trades. I felt good making my dollar profit goal, but I feel conflicted because I didn't let my profits run.

Which is the most correct course of action?

A: I think you need to do some conflict resolution. First, decide who you are? What are your beliefs about yourself? What part of you wants to let profits run and cut losses short and what part of you wants a certain profit goal each week?

Next you need to find the positive intention of each part and do a conflict resolution between the two parts. We do those sorts of exercises in the Peak Performance workshop and they are also in Volume 3 of the Peak Performance Home study course. When you sort out the conflict, then you will have the best answer for you.

Posted by Van K Tharp at 02:13 PM | Permalink | Comments (0) | TrackBacks (0)

Q: I have purchased your home study course, 2 of your latest books and your special report on position sizing. Position sizing has had the most impact on my trading. I still think I am missing something. I may be transferring a fear of letting my profits run into a position sizing scenario. If I can anchor my emotions and thoughts to a specific "rule of thumb, " I can trade much better.

When I place my 2nd entry after a profitable 1st entry, I always have a stop on the 2nd entry.

If the 2ntry stop is hit. What is a "rule of thumb" to do with the 1st entry? What are the parameters
for the trailing stop of the 1st position?

1) close it out at the same time as the 2nd entry stop loss.
2) close it out at break even point for both trades.
3) close it out at my 1st entry fill price ( a 1R loss + commissions)
4) close it out at original stop loss ( a 2R loss + commissions)
5) Do not close 2nd entry at regular stop price, close at breakeven point of trade.
6) Other

My guess is that it varies for different systems. But I am looking for some type of parameter.

A: I probably won’t directly answer your question, but will instead ask you more questions. Have you done the following? I’m asking them, because your question suggests that you have not done so.

1) Who are you? Determine your beliefs about yourself, your strengths and weakness, your edges, etc.

2) Once you’ve answered that question, you can determine your objectives. What are you trying to accomplish as a trader? Do you know that? The answer is 50% of system development.

3) What are your beliefs about the market? You can only trade your beliefs. Did those beliefs come from you? How do you know they are useful? Do they limit you in any way?

4) How good is your system and how does it perform in the 6 market types? My guess is that less than 0.001% of all traders have answered this question.

5) How can you use position sizing to meet your objectives? This is probably one of the most important questions you can ask yourself.

The answers you are asking for are basically “It depends…..!” But if you can answer the five questions I have given you, then you’ll probably also find the answers to the questions you have asked.

Posted by Van K Tharp at 09:26 PM | Permalink | Comments (0) | TrackBacks (0)