note to dan - update top123 factors spreadsheet

Research
1

hey dan,

i wouldnt ask this but i remember you saying this was a fairly effortless update. . .is there any way you could post a new/updated list of individual factors performance spreadsheet?

i would really appreciate it. . .and it will five me something to do other than watch all the red lately. . .

2

bump

3

budonk,
Iā€™ll run it soon - Iā€™m down to 1 PC and Iā€™m using that one for other stuff for the next few days. Are you in a hurry ie do you have some time available now to create ranking systems that you wont have next week?

Single factor testing tool developed using the API
4

no hurry at all, and thanks so much! ! !

5

budonk and all,
I gathered the data for the single factor ranking systems for 3/31/01 to 6/17/06. The file is attached.

Port123 factors list-June06 shared.xls (948 KB)

1 Like
6

Dan, youā€™re the MAN ā€¦ enough said!

Thanks

7

Dan, thanks so much for providing this! Two questions:

  1. What software are you using to get this data?
  2. Do you have any rules of thumb on updating ranking systems on this data? I use the TopFactors ranking system for one of my ports and wonder if I should modify it based on the updated spreadsheet.

Thanks.

8

areid,
Your welcome. Iā€™m glad you find it useful.

I use an automated software testing tool to gather the data. There was a thread on this in the past.

As far as updating ranking systems, I personally have one port that is based on the ranking system we created as a group back at the beginning of the year. The Ranking System name was ā€œTF12-05 ExcellentOnly-Optimizedā€. There are several public copies of it out there. Iā€™ve been pretty busy but once I have some time I plan to experiment with the top factor results from June to see if I can improve the ranking system.
A good test for the older ranking systems is to run the ranking system performance charts from the time you created the ranking system to the present. This period is ā€˜out of sampleā€™ and will give you a good idea if the ranking system is really good or if it was an example of curve fitting.