New Data Base

[Dino D.]

Hi everyone, just did a quick check of my new database cross checking with my "trading profile" that all my codes are upto date!

The new database seems to have quite a few stock codes missing for the ASE, when I select all codes and export them to excel I only get 479 ASE codes.

I think there is supposed to be over 500?

What does everyone else have?

Dino.

[Mark Linton]

There are 462 different stocks that have given at least 1 trade in the latest research database.

[John Kennedy]

Hi Mark

Any reason for the dramatic reduction in the number of trades now included in the new database compared with the old one.

ASE Equities - Old - 18863  New - 16788

ASE Sectors - Old -    751  New -    695

USA Equities - Old - 65201  New - 35020

Is it that we have lost a lot of valuable history and past research or is it that a very large number of stocks have been deleted, particularly for the USA.

We have already lost the what was valuable Short trade history would hate to see it for Long trades.

[Dino D.]

John, I have the same number of trades in my new database, my concern is the list of stocks inside the Research Tool that we select to make up our trading profile.

Last time I updated the database in June 07 I was trading with an incomplete list of stocks & it wasn't until 3 months later I realized my database was corrupt! (a heap of codes that were in my trading profile simply were not in the database)

When I called Sharefinder & compared Mark's database with mine I realized I was missing about 80 stock codes, I ended up re-installing the June 07 database which fixed it.

John, if you select all the stock codes from the database inside the Osprey Tool>right click>select copy>open excel & paste, how many codes do you have for the ASE database?

Dino.

[John Kennedy]

Hi Dino

Answer is I don't know.  I am waiting a week before uploading to make sure any initial gremlins have been picked up.

No doubt I am going to have to redo all my past research if the database has changed as much as it appears to. 

That will take some time so I will need to find a slot when I can find the time to put in the best part of a week or more of solid work and I just don't have the free time now days.

I just hope the trades database has not changed too much but given the numbers and lack of any detailed explanation I suspect not.  Last time they did it it to us I found it very frustrating as it meant most of my carefully researched and kept records were basically useless.

My problem is when I look at the outcomes from a broad picture on a very large spreadsheet tabulation the outcomes appear to be pretty random.  There are no obviously discernable patterns, probably need to do a lare scale multi variate factor or similar analysis on the results but no longer have access to the needed powerful statistical analysis software or the interest in taking it that far and also I doubt if that would show up any useable correlations, eyeballing is a pretty good broad brush tool. 

In doing it last time I tested a range of reasonable and maybe unreasonable scenarios going back in short or long periods and had difficulty in identifying a tradeable pattern to it all, whatever the SF research results show. 

To try to better reflect the real world I pick a past period and project it forward for the next 12 months, as all good stock market research says you should, and it did not look particularly good when I did that.   

I am also not over enamoured of Version II with its much higher drawdown risk.  Some of my models were reaching very high drawdown numbers before they turned around or you had to reduce position size to well below what I considered reasonable given my capital base.  Any patterns I did find when projected forward did not show sufficient promise for me.  However, compromises being what they are I was prepared to put risk Capital against the results I came up with. 

In the end I used a maximum database including all past trading history which then worked in most years but I was still not really happy with it.  It seemed to me to give better overall results than shorter time frames.  Maybe did not maximise returns but evened out returns.  My practicle experience was that looking for maximised returns did not work, I found it difficult to enter the good trades and even more difficult to exit with profits intact, market depth is so critical in AUS.  This was compounded in Version II where I think there is much more reliance on a few good trades to generate outperformance. 

My conclusion is that version II is probably more suitable to people with a low Capital Base who are younger than I am and are prepared to take on more unanticipated risk. 

I now have to test out my past research with the new database to see if my last selection of trades would have given the returns I need to justify the time and effort involved.

Contrary to that though I have generally made money out of Osprey by generally following the rules as close as I can and always 100% with my USA trading.  For Aus my best results though were always when I was able to put in time and effort to actively manage the trades, but have not now traded AUS or USA since June last year.

John

[James Oliver]

Hi John,

I have read your comments with interest, since I also have struggled to identify a tradeable trend (particularly when projected forward) ... past trading history, currently appears to be a poor indication of future returns with large swings in profitability, high drawdown & varying expectancies.  Practical experience has also demonstrated that gapping and slippage on both entry and exit have made CFD's unattractive for automatic entries (i.e. trading with buy and sell stops).

In the end, I found that trading SPA3 with leverage provided greater consistency and consumed far less time than persisting with Osprey.  I have also found that the returns (whilst not as high as Osprey due to the lower turnover) are also very good and the drawdown was acceptable.

James.

[Dino D.]

Well John & James, thanks for your reply & you have pretty much hit the nail on the head there.

To sum it up in short, I've pretty much stopped trading Osprey in Nov after my V2 portfolio went into meltdown.

I don't think I will resume trading V2 or V1 Osprey simply because (in my opinion) both systems were not profitable in the current market. It takes so much effort not only with the research, but also the daily choirs of trading a system like this.

I never traded the US market but I found liquidity was a problem with ASE(increasing slippage) & profits werent large enought to take me out of drawdown.

I only updated the database out of curiosity on how the portfolio would have ended up if I kept trading. And it doesnt look good!

Talk soon.

Dino.

[John Kennedy]

Hi Dino

I have now downloaded the latest database.

Initial results if traded all last 12 months ie from 1 Jan 07 would currently be sitting on over 60% drawdown on Aus and 87% on USA.  Just confirms my suspicions.  Both charts have been all downhill since late May last year.  Also looked at putting in my researched trade database for Aus from 1 July 07 and the same picture although flattened out over the last 6 weeks or so of data.

Have not checked stock selection to see which ones if any are no longer included but there does not seem to be any point.

In the current market Osprey is in my view non-tradeable.

John