Klein Decisions Archives
The KleinPost

Feature Article

Inside K4

Model of the Quarter

Tips from the Klein Bottle

Ask Felix

The ReKleiner


More Ask Felix

A-Share Filter

Finding Your Factors

Painless Profiles

The Disappearing Fund

Expert Export

A Case of Acute Elimination

 
Click here to see Ask Felix.
 

Sign Up Today!
* required
*

*


*

*


Powered by VerticalResponse

 

68 T.W. Alexander Drive

P.O. Box 13628

Research Triangle Park, NC 27709

919.233.6767

 

 

 

 

 

 

 

 

 

 

 

 

A Case of Acute Elimination

By Dr. Felix Kleinstein

Question from respected colleague:

When I evaluate domestic equity funds, I:

  1. Eliminate all that have loads or 12b-1 fees

  2. Eliminate those that are closed to new investors

  3. Require at least a 3-year history

  4. Eliminate those with less than .90 R2 to the style over the past 5 years

  5. Eliminate those whose returns fall below the 50th percentile in their style over the past 5 years

  6. Eliminate those with a negative alpha for rolling 3-year periods over the past 5 years

  7. Eliminate those whose Standard Deviation is above the 50th percentile of their style over the past 5 years

Can I do this with Klein K4 Fund Selection?


Answer from Felix:

Actually you can do a whole lot better. You see, my friend, you have an acute elimination problem. No, no, don’t go running to the pharmacy: you simply need to change your way of thinking. You’re obviously still working in the world of elimination screens where you set a number of hurdles and find which funds survive them all. Some funds may fail miserably on all counts while others may miss just one criterion by a minimal amount. You have no way of knowing. Worse yet, since each of your tests is a pass/fail ultimatum, you have no way to distinguish those that are more important. They’re all equally important. As a result, the funds that make it through your screens aren’t necessarily the best funds, they’re just the survivors. Not only that, all you have is a list of funds; you still don’t know how they compare relative to one another. Ah, the symptoms of acute elimination.

Fortunately, K4 Fund Selection is the cure for your problem. With K4 everything isn’t pass or fail, there are degrees of acceptability and levels of importance, and all funds are ranked on them. Sure, there are some features you want in all funds. K4 treats these “must-have” traits as filters – pass/fail tests you apply either before or after you’ve run your evaluation.

The remainder of your criteria are “preference items”. Rather than stating a cutoff point to eliminate funds, K4 allows you to set levels of importance for each of them and then it ranks them. In addition, you’d probably prefer funds that have positive alphas (item 6), and also those with higher values to those with lower values, right?  You can do that when you aren’t forced into rigid elimination screen thinking. K4 gives funds higher scores if they display more of your preferred characteristics. Isn’t that a lot more intuitive?

Finally, in the results, K4 isn’t going to eliminate funds; rather it’s going to rank them all based on their scores on the combination of attributes. Even if you apply your filters (items 1 and 2) to eliminate load and closed funds, you can always disable them to see the rankings for the entire category. As a result, you don’t just have a random list of survivors; you have a rank ordering starting with those that display the greatest levels of the features you decided are most important.

So, here’s the bottom line for your model:

Category: Funds, Equity, Track Record > 3 Years
Preference Items: 5-Year R2 to Category Index
  5-Year Return +/- Category
  5-Year Rolling Alpha (3 in 5)
  5-Year Relative Standard Deviation
Filters: Maximum 12b-1 fee = 0
  Closed to New Investors = No

Once you do this and see your results, your elimination problem should be cured. Now, go run two scenarios and call me in the morning.