Trader Fred's System questions

[VirginiaBob]

Does the new trading system by "Trader Fred" incorporate TSP tricks that TSPtalk members have figured out such as G-fund penny days and I-fund future values? If not, these would be items to incorporate in the decision making process due to increased/decreased probabilities of gains duing these situations.

 

[tsptalk]

No, but Trader Fred is open to any data that will help returns. I'm sure the system, which was very successful trading Rydex funds, will be an evolving process as data comes forth.

 

[qibovin]

Is there any detailed information on this system that I'm missing, or will it be forthcoming? I read the page about the 8 indicators and if any one is a buy, then the system is a buy. But what are these indicators? In other words, what do A, B, etc. represent or measure and how are they computed? Has this been posted anywhere or is there a link that I'm missing, or am I totally missing something that makes this sound like a really stupid question? (Because otherwise I'm surprised that no one else has asked this yet.)

 

[tsptalk]

You're not missing anything. While he has given me little tidbits of info, data, charts, and stats, the only thing we really have, or will get, is the buy or sell signal, with possible hints as to whether the signal is stable, weakening, or getting stronger. This is one reason he has insisted that I offer a lengthy free monitoring period. Back testing is not enough to warrant a winning system so he wanted it to be in live mode for several months.

As I learn more about the system I can talk more intelligently about it. Right now "Fred" and I talk to different languages (although it's all in english). He sends me those charts and stats and I say, "what the heck does all this mean?" But I'm learning.

The first fund model he wrote made "Somewhere in the double to triple digit multi-million dollar range after expenses" for a company he worked with, in just few years. It was written up in the Wall Street Journal and he was sought by many people who wanted to pick his brain - the last thing he wanted. He has broken off ties with them but he assumes they still use it.

To add, the big money was made both going long ETF type funds, and short during buy signals, something we can't do with TSP, but certainly in our our personal investment accounts.

Also, he has been experimenting with, and having success with, 4 additional submodels.

That may not be want you want to hear but I hope that answered your question. If you have any specific questions, I can always pass it along to Trader Fred and see what he has to say.

Tom

 

[qibovin]

Thanks, Tom.

I can appreciate this, particularly specifics of formulas and such. Still, I'd feel more comfortable with some sort of discriptive, but not particularly revealing, names for these bars, like "moving averages indices" or "sentiment factor." Since this "system" is intended, as I understand it, as just one tool for consideration in each of our personal risk management decision processes, some way to modify the weight of (or ignore) each of these factors seems important or, perhaps, at least, what sort of risk is built into the system...

Well, I guess we'll all learn more as we go. I'm all for making money.

 

[tsptalk]

I'll pass along as much as I can.

Risk seems to be a major factor. All the back testing has shown that we might be in the market about 70% of the time - As oppossed to 100% with buy and hold.

 

[Show-me]

Tom,

Normally we are buy and sell signal people here. Is there no "hold signal" in Fred's system? Will there be one in the future? My point is I do not what to buy at the top and his models are showing some weakness. A "buy" signal at this point in the game may be ...............

 

[tsptalk]

Good point. An actual buy is given when it moves from sell to buy. I have been saying it remains on a buy but I guess it would be the same, or better, to say hold. I will reference the internal buy and sell signals that go on all the time with the individual sub-models while the system is on a buy signal. It won't mean anything except to perhaps see if the signal is weakening or not.

A sell means all sub-models are in sell mode, and you'd "hold" that sell until the next buy signal.

So there is no "hold signal" at the moment in the sense that you would hear an analyst say hold a stock (rather than buy more) but that could be something to work on.

Thanks,
Tom

 

[lsitter]

I read somewhere about what percentages to assign to the C, S, & I funds using the system. Will those percentages always be the same when a buy signal is given to come out of G or will they vary based on the different signals?

If they vary, will the percentages ever change during a particular buy period. For example, we come out of G and go 35 C, 25 S, and 40 I per Trader Fred's recommendation. Then the signals change and indicate that we should go 50 C, 40 S, 10 I. In other words, will we ever be recommended to reallign our percentages during a buy period or do we stay with what we have until we are told to sell?

Thanks, Lou

 

[qibovin]

"The Buy signal means to go from the G Fund to some combination of the C, S, I Funds the very first day the Buy signal is listed on the TSP Talk web site. The official trade is entered only on the first day the Buy signal is listed on the web site and not on any subsequent days. After that first day, the trade enters its Hold mode until a Sell signal is given to exit that trade and return to the G Fund. The trade is always exited on the very first day the Sell signal is listed on the TSP Talk web site. Exiting the trade after the specific Sell day has passed nearly always yields completely unpredictable and possibly catastrophic results.

"Since we never really know in advance, the duration of a Buy or Sell signal, for each trade there is officially only one day to Buy into some combination of the C, S, I Funds and only one day to Sell those C, S, I Funds and return back to the G Fund. Outside of those trade dates, the results are unpredictable."
Trader Fred (from todays Market Comments)

Given our already limited trading opportunities (one trade per day at COB if posted by 1200 EST each day) how much time will we actually have to make this transaction? I presume from your (Tom) daily comments that the buy or sell signal will get to you prior to that morning's Market Comments, usually done NLT 0800 or so. That really leaves us a 4 hr window in which to respond to changes, otherwise results will be "unpredictable." That's a pretty tight window. (Not that I'm complaining.)

 

[tsptalk]

The system signals are determined by the end of day share prices so yes, any buy or sell signal will be in the daily comments. Because of that, the email alert should go out much earlier than our current ones, which are usually sent within the hour of the deadline. Most likely they will be out the night before the deadline.

As we know, one day does not normally make or break a trade, but that is the way Trader Fred thinks. He likes to use the analogy of a rifle shot vs. a shot gun shot. Be as precise as you can. Since the system has been in a buy signal since August, you can see this would have worked out equally as well if you were a day or so late. Of course not all trades will not work out that well.

Tom

 

[tsptalk]

I read somewhere about what percentages to assign to the C, S, & I funds using the system. Will those percentages always be the same when a buy signal is given to come out of G or will they vary based on the different signals?

If they vary, will the percentages ever change during a particular buy period. For example, we come out of G and go 35 C, 25 S, and 40 I per Trader Fred's recommendation. Then the signals change and indicate that we should go 50 C, 40 S, 10 I. In other words, will we ever be recommended to reallign our percentages during a buy period or do we stay with what we have until we are told to sell?

Thanks, Lou

Lou -
That is still in the works. Fred has been working on using the sub-models within each fund to determine strength which would help us allocate. Currently there is only one model which is looking at the S&P 500 to determine the overall buy/sell status. He will test that to make a determination if it is feasible or not. Until then it will be up to us to decide between the stock funds. He has given that spread recommendation on www.tsptalk.com/system.

 

[lsitter]

And just to make a long story longer... That spread that he gave us: Although currently it will remain the same for the duration of the buy period, will it be a different spread for a new buy period, one that comes after the sell period? Or, will all buy periods be 25/40/35?

Thanks, Lou

 

[tsptalk]

At this time, we will use 25/40/35 based on his recommendation and backtesting results. Although I may provide some input, as to what fund might be stronger going forward and why (interest rates, the dollar, etc). Basically the buy signal says go ahead and get aggressive, as aggressive as you'd like t get.

During the backtesting the system was in stocks 70% of the time, and in the G fund the other 30%. So I'd say take advantage of that 70% by being aggressive, but as the sub-models start giving individual sell signals, and we'll tell you that, you may want to play with the spread a little or even back out some if you'd prefer. That will be up to you. At least until we come up with something a little more codified.

Tom

 

[fabijo]

I'd just like to send some thanks out to Trader Fred. My only concern is that we are only using data from 3 1/2 years ago till now - where the market has been pretty consistent. Shouldn't the testing also be done on different types of markets? There is plenty of data on all kinds of markets to be freely downloaded. For example the Dow Jones Industrial Average gives you data from 1928. That's plenty of market personalities to play with.

Just my 2 cents.

 

[tsptalk]

That should be a concern only in that no one should jump in without seeing solid evidence. The thing is, only the TSP data goes back that 3 1/2 years. The system has been backtested going back much further, and has, in one version or another, been working using other vehicles (ETF and mutual type funds) since the mid to late 1990's. I don't know all of the details, but I know much of the data the system uses was not around many years ago. It's not just share price but several indicators. I'll see if I can get more info.

 

[fabijo]

That should be a concern only in that no one should jump in without seeing solid evidence. The thing is, only the TSP data goes back that 3 1/2 years. The system has been backtested going back much further, and has, in one version or another, been working using other vehicles (ETF and mutual type funds) since the mid to late 1990's. I don't know all of the details, but I know much of the data the system uses was not around many years ago. It's not just share price but several indicators. I'll see if I can get more info.

True true. I struggled with the 3 1/2 year data problem myself. I just figured that a market is a market. But you do say that it has been working with other vehicles, so that is definitely a good thing. Like you said, we just have to watch it in action and we can decide for ourselves. If we get a really bad year and the system saves us from it, then it is worth our close attention.

 

[tsptalk]

From Trader Fred:

If anyone has access to the historical NYSE volume data (see red ellipse) in electronic format going back as far as possible that would be much appreciated. As Tom mentioned, the model uses data elements not around prior to the mid-1990s. However, rational substitutions for increased back testing are possible, but not without the data in the red ellipse in electronic format.

 

[sugarandspice]

With systems there are always "Things" that trigger other things and mean other things. So regarding Fred's comments today..... "A market pile-up occurs when four or five black diamonds are within approximately fifty Loss---Gain units of each other. This time only three black diamonds clustered together, which is not enough to create a market pile-up (dotted green circle)."

Could you ask Fred a question Tom? The part where it says "A market pile-up occurs when four or five black diamonds are within approximately fifty Loss---Gain units of each other." Could this read ....At least four black diamonds....instead of four or five black diamonds? I'm just wondering that maybe more that 5 black diamonds means the system reads something else.

 

[tsptalk]

Sure, I'll ask.