0:00

[MUSIC]

Okay, so in this last video,

we're going to have a look at two additional tools which you may use to find

the best performing managers adjusted to some kind of a risk measure.

And actually, this is one of my preferred measures, and

I'll explain you why I quite like this measurement.

This risk adjusted performance ratio, which is not as common as Sharpe or

Trainor, but in my view, has something very right to it.

Okay, so what we'll do, we'll talk about the MAR ratio.

And MAR comes from a newsletter, which has been around for many, many years.

And it actually says Manager Report Newsletter, and

it has an emphasis on hedge fund analysis, or

indeed, exactly what we're discussing here, peer group analysis.

1:28

And the drawdown to me, and this is why I like this ratio,

it's actually a notion of risk, which makes probably

more sense than is intuitively appealing to a customer.

Who hasn't gone maybe to university and

does not really know what a standard deviation is.

And may have difficulty in grasping it, especially so if that standard

deviation captures a return which exceeds the average.

This may be somewhat difficult for somebody to associate that to risk.

But certainly, a notion that is more straightforward to be

identified with risk is the notion of maximum loss or

maximum drawdown, or also pick to through.

Basically, you say the worst you could lose by buying this security or

this investment fund is so much, and this, to me, has a very clear meaning.

Okay, so a MAR ratio in excess of 2 will be

a manager who delivers maybe 92% total

return since inception, as we see here,

and has had a maximum loss of 46%, and

so 96 divided by 46 is 2.

And just to highlight, again, you have seen this chart already,

but what the drawdown means, it's the worst possible

scenario of entering the market at the peak here in 2000 and

exiting the market when the fear is at its maximum in 2002.

And there, you lose 46%.

And this will be your drawdown, okay.

So, using the MAR ratio to identify,

especially when you have no benchmark in terms of markets,

market indices, may prove a very useful measure.

And indeed, also something which is simple to grasp,

you compare the total returns since inception, and

you divide it by the maximum loss, simple and intuitively appealing.

But now there's a shortcoming to this MAR ratio.

And I want to illustrate that with the following examples.

Assume we have two managers, Manager A as a CAGR,

that's compounded annual growth rate.

Total of 40%, so since inception, and

a drawdown of maximum loss of 20, so a MAR ratio of 2.

Manager B has slightly above total return of 50%,

but also, our shopper pick the draw of loss and

drawdown of 30%, so 50 divided by 30 is 1.66666.

Okay, so the first manager, Manager A, has a higher MAR ratio,

so you would be tempted, you would be inclined to allocate

all your investment to him and none to Manager B.

4:41

But what if I now tell you that Manager B has been around for

20 years and Manager A only for 5 years?

So, clearly, you'd say well, Manager B has more experience and

maybe he has a greater drawdown because that

drawdown happened when Manager A did not exist.

If Manager A had existed, then maybe he would have experienced a greater loss.

So the way to solve the shortcoming of the MAR ratio is the CALMAR ratio.

And the CALMAR ratio is quite simply it's the same notion of the MAR ratio,

but here, instead of looking at the worst

possible loss over since inception, basically here,

you're looking at the worst loss over the last 36 months, so the last three years.

So CALMAR and MAR ratio need to be used together because

maybe the idea of a maximum loss of 5% within the last

three years does not give you an accurate picture.

Because maybe the fund is actually far more riskier than that,

and it has experience of 46% loss 15 years ago.

So you need to use both.

But at least you can be sure that using the CALMAR ratio,

you're comparing apples with apples.

And you're comparing managers who have been in existence and

over the same time periods.

And clearly, this is something that is, obviously, very important to mention.

When you're doing this filtering, when you're extracting from the database,

managers, based on return statistics, risk measurements,

percentage of positive months, and etc, etc.

You need to make sure that you're comparing apples with apples and

that all the managers have a similar mandate when investing their funds.

So in conclusion of this set of two videos,

we've seen here that to perform the peer group analysis,

what you need is a good database that will have managers,

which are ranked by categories.

And we need to make sure that this is done professionally and

each category you do find managers that have a specific investment philosophy.

And then basically, you can filter that database using criterias,

which best suit your investment philosophy.

I give you one example.

If you want to assemble a fund to funds, which will be low risk,

then you will put a lot of emphasis on extracting from your database

managers who have the highest proportion of positive months,

who have the minimum drawdown, who have the lowest volatility.

8:00

then you may be just shooting for the maximum beta, the people who use leverage,

and maximum return strategy, you will put more emphasis on return.

So this all depends on your philosophy, and

then in the second video, we saw a measurement, which in my view,

is very useful when we are talking about peer group analysis.

And we don't have a benchmark to compare ourselves to.

So basically, here, we're just looking at a total return,

dividing that total return by the maximum loss.

And this is, to me, a very intuitively appealing notion of risk.

[SOUND]