Our Indicator Current Value

[sawyer]

I was impressed by the way the MTI alert level followed the market down, while giving a sell signal on the market. With the buy signal since the Dec. 29th, the alert level has not changed very much. When the TSX closed, very close to 9500 on Jan 6th the spread was very large. (close to 1000 points) So generally, your alert level seems to act differently depending on whether there is a buy or sell on the market. When I look back at the TSX to the year 2000, staying out of the market when below the 50 day MA, appears like it would work well. I wonder how your portfolios would have performed using that simple principal.

[Super Stock Picker]

Hello sawyer,

Well, in its definition, there's nothing different for UP and DOWN signal. I think what you've seen is more about the market behaviour itself.
If the market keeps going down for a few more days, it would make the last rebound very short lived. If you compared to the last DOWN signal, the market has kept the same direction for a while and with an impressive pace.

Regarding the years 2000, our portfolios were not there yet. So we can not back test this period. We can back test the indicator itself, but as its main usage is the volatility reduction of our portfolios, it might not show its full effect when only the market as a whole is considered.

Regarding a 50 day MA, it is obviously a good indicator of the medium term trend. But when building an indicator, you have to consider how efficient it is to give you an information about the current trend and also how often you could be whipsawed.

[garilou]

Hi SSP,

Could you elaborate a little bit more on what you wrote to sawyer, when you say:

We can back test the indicator itself, ...<snip>...it might not show its full effect when only the market as a whole is considered.

Is it unhiding the "recipe" to tell us what else as the market as a whole is considered?

Thanks,

Louise

[Super Stock Picker]

Hello Louise,

No, sorry, we were not giving any clue about our recipe.

What I was mentioning is that we could back test how the indicator would have timed the market. Then, we can compare this timing to the market without timing.
But this is not how our indicator should be assessed. Its purpose is to lower the volatility of our portfolios. And as our portfolios have no history before 2004, the back test of the indicator before that date would not bring a lot of information.

Does that explain better the meaning of our previous post?

[jmath38]

It would be interesting that you update the graphs of the UPv4 with and without the market timing indicator to show the work of this indicator (the graph in the "home page" of the tab "Market timing").

Thanks,

jmath38

[garilou]

Hi SSP,

No, sorry, we were not giving any clue about our recipe.

It's OK!
This why I asked before, if elaborating would mean unhiding the "recipe", because I know you do not want to give the recipe.
I was not asking for the recipe.
For the rest, I understood pretty well the meaning of your previous posts.

But as jmath suggested, it might be interesting to see what happened to the UMPv4 since the DOWN signal was lifted, as you had done before, while the DOWN signal was still active, and a comparison with and without the MTL.

But I would not insist on this, since you said recently that the tool is not completed, that you are still working on it, and you might think again that I am trying to get the recipe, which I am not!
In between I am trying to develop my own little one... and it is far from easy!

Louise

[jeff]

from 29-Dec-08's up signal to to today's (20-Jan-09) down signal , we came back to where we were in December. So December 29th's up signal did not generate any gains. 

[garilou]

Jeff, you write:

So December 29th's up signal did not generate any gains.

We are in a difficult period, that's for sure, but I think you are somewhat too negative.
Yes it's sad to be back where we were, in a sense, but to say that the UP signal did not generate any gains is not quite true.

During that period, and even before the previous DOWN  signal was lifted, many SSP choices made pretty well.
In another topic, we discussed a lot about DDS, but it generated a pretty good profit for those who were not turned away by the E-Trade false alarm.

SXC did a pretty good ride too, although a good part of it was during the DOWN period.
As we understood, the goal of the MTL is to prevent too much volatility in the portfolios.

Even if we do not know the MTL recipe, the buy and sell orders still come in, which means that some stocks still have (or had) some momentum in them.

Whether "we" make gains or losses, we are in among the lucky ones who can afford to trade on the markets.

The really sad thing about that DOWN signal, is that it reflects that lots of people will have a real hard time in the coming months.
 
Louise

[jeff]

Hi Louise,
 
after the December up signal TSX did rise a lot, but if we followed the signal strictly and did not sell TSX until the down signal was issued, we gave back all the TSX gains and ended where we were. with my last post i mean: This completed up-down cycle did not generate gains.

[Super Stock Picker]

It would be interesting that you update the graphs of the UPv4 with and without the market timing indicator to show the work of this indicator (the graph in the "home page" of the tab "Market timing").

Thanks,

jmath38

Hello jmath38,

This is done.

[Super Stock Picker]

Hi Louise,
 
after the December up signal TSX did rise a lot, but if we followed the signal strictly and did not sell TSX until the down signal was issued, we gave back all the TSX gains and ended where we were. with my last post i mean: This completed up-down cycle did not generate gains.

Hello jeff,

Do not forget that this is not the primary purpose of this timing indicator.

We do not assess the indicator in terms of points gained over the market or by counting a success percentage of good calls, but we want it to give us a timing to reduce the volatility of our portfolios.

The indicator, by itself, is not that good, as it does add value to the market itself (except in 2008 of course). To make it useful, you have to time something that is more volatile than the market.

Our portfolios are good examples. You can also think at timing some ETFs that give 2x the market perf.

By the way, do not forget that there will be some wrong calls and whipsaws as well along the way. You have to expect this.

[garilou]

Hi SSP, Hi Jeff

SSP, you have updated the image showing the influence of the Market Timing Indicator, but since you go back so far in the time, it is somewhat difficult to see the real effect for us who had the first information as to those UP and DOWN signal on October 24 2008.

In order to show more the influence this tool can have to protect the portfolios, I have prepared this spreadsheet, taking all the orders (closed and still opened as of Jan 21), that were in the range from October 24 2008 till Jan 21 2009 for the portfolio UPM v4.
I think this will be, because closer to now, easier to see how the MT has changed the results.
Sure they are somewhat theoretical, because I took, like you do, the closing prices, which could differ slightly when one bought or sell, not only because the price fluctuates during the trading hours, but also because some were hard to buy or sell because of the volume problems (but this is another topic).

And Jeff, it might show you that it might be true that the portfolio did not gain much during those UP and DOWN signals, but that we could have lost much more! Weather the indicator is perfect or not, it worked!

And thanks again, SSP, to have us let us known about the indicator before it was finished.

I hope this will be useful, like a magnifying glass.

Louise

PS. I have not taken into account the allocation system. I just compared opening and closing positions. The results could be different, but I think not many users are strict followers of the allocation and re-allocation system. And since my %age results are the same as those posted by SSP, it  lets me think that SSP does not apply it's allocation system either  .

[Super Stock Picker]

Hello Louise,

About the graph, you can also right-click on it and select "view image". That opens the image bigger, so it is a bit easier to read.

A note about your research: you have been able to check the average return of the recommendation with and without timing. The difference between the two results shows that the timing has helped over this period. But, the figures are different from the portfolio's return over this period.

That leads to another part of your post. We are still using, and we will always do, the same allocation rule as ever. It is all automatic so it will never change. The fact that you can find the same returns without our allocation is certainly due to the fact that no stock has got a huge positive or negative return during this period. As a result, the portfolios performances are not too far from a simple average.

[garilou]

Hi SSP

About the graph, you can also right-click on it and select "view image". That opens the image bigger, so it is a bit easier to read.

You can imagine that I had done that first!
But big or small, the picture still showed so many years back that the short period between October 24th and now were still very difficult to see.

When I wrote the post I was pretty tired, because making that spreadsheet took me a lot of time.
It was my last interpretation that was wrong when I wrote:"...it  lets me think that SSP does not apply it's allocation system either" ,
I am somewhat mad at my-self that I did not think long enough to see that sentence was really erroneous.

But you also did not see where the mistake was.

the fact that no stock has got a huge positive or negative return during this period.

The real reason?

I took the data for each single stock, and compared the percentages using the history page, where you show the buy and sell orders, with the dates and close prices.
On that page we only see the progression stock by stock, and not the progression of the portfolio with the allocation system.
That is why I come to the exact decimal to what the history page showed.

When you post your portfolio returns, you do apply your system, but those are on another page.
Doing the same exercise applying the allocation system would have been in retrospect more exact, but I would have needed days to do that..

Still I think that the spreadsheet shows that without the Market Timing, we could have lost much more, which was my intent.

I just hope that I did not introduce confusion for those who would try to do as I and a few other people did: take the time to really understand you allocation system.

I have "paper tested" it a lot, and could see that using the allocation et reallocations really brought better returns then simply buying and selling some of the stocks recommended. Sure, "paper trading" is never the same... still it is worth applying it, (in better markets as now).
But as long as the volume filter will not be completed, it will stay hard to apply you system: when the allocation said: buy 8000 shares of a penny stock, and a few days later, sell 1 or 2 thousands in order to buy another one or to ajust for the new prices of the stocks in the portfolio, when that stock traded in a way that it was almost impossible to buy or sell it, at least at prices comparable to your system that only considers the closing price, and that even if only 1000 shares or sometimes less were really traded on that day, or one big trade had brought the stock 5 to 10% higher or lower...it is much easier on "paper" than in real life.

But (repeat) this is another story, I guess it will come sometime.

Louise