Passive or Active Fund choice

arabianights

dunstonh wrote: »

I cannot understand how you make a consistent 4th-6th percentile fund appear in the top percentile of it's sector. If that means I have a limited grasp of stats then so be it.

Rather than be rude about it, why not explain how you magic the fund up the tables as I am sure I am not the only who would be interested in that?

This is primary school stuff !!!!!!.

Let's keep things *really* simple... Three funds A,B,C and 3 years starting from a base of 100.

Year 1: A 130, B 120, C 90
Year 2: A 135, B 140, C 120
Year 3: A 145, B 160, C 145

So B increases the most despite the fact it's in the middle every year...

dunstonh

So B increases the most despite the fact it's in the middle every year...

Are you a troll?

Lets take the YTD performance using the stats mentioned earlier. (YTD: 192 / 375)

According to you the fund that had the 192nd most growth this year (out of 375) was actually in the top 37 for growth and not really 192nd.

What about the fund that came 191st or 90th or 80th or any of the 191 funds that had better performance?

arabianights

dunstonh wrote: »

Are you a troll?

Lets take the YTD performance using the stats mentioned earlier. (YTD: 192 / 375)

According to you the fund that had the 192nd most growth this year (out of 375) was actually in the top 37 for growth and not really 192nd.

What about the fund that came 191st or 90th or 80th or any of the 191 funds that had better performance?

:rotfl:

You still aren't getting it...

What relationship is there between the top performers one year and the top performers the next?

Answer: None has been found.

arabianights

This is something I feel very strongly about in fact... so I've got the OP another link:

http://www.economist.com/world/britain/displaystory.cfm?story_id=E1_TNNQVTP

Retired_I.F.A.

The only thing worth reading from that link is this..

..he found, would have to put in £1.55 to get a return equivalent to £1 invested directly in the market.

Why? , because it's absolute crap funnier than any cartoon I've ever seen. The American Economist who wrote that ought to go back to school and learn basic maths or do as his grandpappy did.... jump out of his 30th floor skyscraper window.

Retired_I.F.A.

Simple common sense...

1/ look at past performance tables of all funds

2/ decide on what portion you want in what sector(excluding trackers)

3/ Pick the funds that show consistent 1st or 2nd quartile performance over 1, 3, and 5 yrs

4/ Do the same again on a regular basis and switch if better are found.

or pick tracker funds which by their very make up will always be middle of the sector.

jamesd

davidmt83, the job of active managers is to go up more than what they are benchmarked against and go down less. Some humans and teams are better at doing that than others. Studies usually completely ignore changes of (human) fund manager even though that's what generally makes the difference between doing consistently better than average and doing consistently worse than average.

Unfortunately I don't know enough about the performance of your two fund options to know which to suggest. Given lack of information with which to determine that the actively managed fund has a better performance record, the passively managed one, if a tracker, is lower risk because it's less likely to be managed by poor managers. With active management the active managers can be good or bad and you need to know which.

arabianknights, do you also believe that the efficient markets hypothesis is true in the real world and for UK investors in UK authorised collective investments?

arabianights wrote: »

What relationship is there between the top performers one year and the top performers the next?

Answer: None has been found.

This claim that no relationship has been found is very easy to disprove and I did so, using real data on relative performance:

Here's how the top 10 global funds in 2003 ranks did for 2003 through 2006:

Neptune MM: 1 2 6 105 New manager 8/05
Invesco: 2 14 63 14
First State: 3 101 2 173 New manager 8/06
SJP Recovery: 4 50 39 34
GAM Global: 5 68 64 24
M&G Basics: 6 1 5 16
Jupiter: 7 4 10 7
M&G Int: 8 5 61 46
New Star: 9 72 3 133
Neptune Global A: 10 9 1 5

It's obvious that for that time period, selecting a fund based on past performance was a good indication that it was going to continue to perform well compared to others in its sector, unless there was a manager change, which is usually a trigger for moving to a different fund or at least considering doing so.

What this doesn't say anything about is how this sector performed compared to others - it's solely about relative performance within the sector. It also says much less about what happens to the relative performance of the funds if market conditions change and this is important because managers can often do well in one type of market conditions and badly in another. The period above was one of relatively consistent market conditions.

Given your area of investment activity I expect that you don't believe, if you ever did, that events five standard deviations from normal are uncommon once you have enough possible events in the statistical universe but I'd be interested in knowing if you do believe it.

arabianights

jamesd wrote: »

Here's how the top 10 global funds in 2003 ranks did for 2003 through 2006:

Neptune MM: 1 2 6 105 New manager 8/05
Invesco: 2 14 63 14
First State: 3 101 2 173 New manager 8/06
SJP Recovery: 4 50 39 34
GAM Global: 5 68 64 24
M&G Basics: 6 1 5 16
Jupiter: 7 4 10 7
M&G Int: 8 5 61 46
New Star: 9 72 3 133
Neptune Global A: 10 9 1 5

It's obvious that for that time period, selecting a fund based on past performance was a good indication that it was going to continue to perform well compared to others in its sector, unless there was a manager change, which is usually a trigger for moving to a different fund or at least considering doing so.

What this doesn't say anything about is how this sector performed compared to others - it's solely about relative performance within the sector. It also says much less about what happens to the relative performance of the funds if market conditions change and this is important because managers can often do well in one type of market conditions and badly in another. The period above was one of relatively consistent market conditions.

I don't see any pattern there at all... and you can't ignore a change in managers because you don't know that's going to happen when you get into the fund :rotfl:

Show me some evidence that past record is a consistent predictor of returns (or at least nearly so.... shouldnt' be hard to some backtesting) and we'll talk

jamesd wrote: »

Given your area of investment activity I expect that you don't believe, if you ever did, that events five standard deviations from normal are uncommon once you have enough possible events in the statistical universe but I'd be interested in knowing if you do believe it.

What's this got to do with anything? Standard deviation only makes sense in terms of the normal distribution, which is not a feature of EMH per se - it came in as a way of modelling random walks... and we all know how that turned out (if you don't ask Mr. Meriwether).

My arguement doesn't rely even on weak form EMH though.

arabianights

Retired_I.F.A. wrote: »

Simple common sense...

1/ look at past performance tables of all funds

2/ decide on what portion you want in what sector(excluding trackers)

3/ Pick the funds that show consistent 1st or 2nd quartile performance over 1, 3, and 5 yrs

4/ Do the same again on a regular basis and switch if better are found.

or pick tracker funds which by their very make up will always be middle of the sector.

Common sense?

Can you backup the common sense experimentally? It really wouldn't be hard to do...

Retired_I.F.A.

arabianights wrote: »

Common sense?

Can you backup the common sense experimentally? It really wouldn't be hard to do...

I retired 5 years ago at 45 will that do?