Excellent points. I love this Group!
--- In [email protected], "wavemechanic" <fim...@...> wrote:
>
> For the past several years, ILF and MALTX are usually highly correlated with
> a correlation coefficient hanging around 1 but before then things were not so
> clean and there were periods of quite poor correlation. However,
> irrespective of their correlation their portfolios are quite a bit different
> so they are really apples and oranges. ILF is fixed at the S&P LA 40 Index
> and MALTX is whatever the manager feels like doing on any particular day but
> typically has about 100 positions in the fund with an average turnover rate
> of "X" days. As a result, their relative performance will sometimes be very
> close and less so at other times. The relative performance (%) for various
> periods picked at random are as follows (starting date shown to present):
>
> MALTX ILF
> 8/20/03 480 398
> 4/12/05 200 178
> 7/05/06 76 70
> 6/07/07 13 14
> 2/15/08 -4 -7
> 8/21/08 1.4 -4
> 9/14/09 11 9
>
> How they behave with any particular system is a different matter and may or
> may not be the same as the relative performance data which by eye appears to
> give the edge to the fund.
>
> Bill
>
>
>
>
>
>
> ----- Original Message -----
> From: Sidney Kaiser
> To: [email protected]
> Sent: February 24, 2010 10:01 PM
> Subject: [amibroker] Mutual funds are outperforming ETFs big time, but they
> shouldn't...why?
>
>
>
>
> I ran a couple of quick tests to check out this possibility. I used
> ILF and MALTX from 1 Jan 2006 to date.
>
> 2 ma xover: the fund outperformed etf by a large amount but both
> grossly under performed B&H.
>
> Kirshenbaum Bands: the etf return was quite a bit greater than the
> fund and both considerably out performed B&H.
> etf 212% 76% b&h
> fund 140% 61% b&h
>
>
> For this test I did a quick optimization of the system parameters and
> then ran the old V4.4 backtester to generate comparable results for the two
> securities on one run. Running the same system with the same parameters over
> the same time span.
>
> Perhaps you might want to keep poking around with this idea.
> Cheers
> Sid
>
