[R-SIG-Finance] quantstrat - Guy Yollin blotter 2014 presentation

[amarjit chandhial]

Brian,


If you regard this is to be poor etiquette, I'm sorry to you and anybody else who finds this to be poor etiquette.


Amarjit



----Original message----
>From : brian at braverock.com
Date : 11/09/2014 - 17:07 (GMTST)
To : r-sig-finance at r-project.org
Subject : Re: [R-SIG-Finance] quantstrat - Guy Yollin blotter 2014	presentation

Amarjit,

Prof. Yollin has addressed your poor etiquette.

I'll address the substantive portion of your email.

On 09/11/2014 01:45 AM, amarjit chandhial wrote:
> In Slide 56 charts.PerformanceSummary(returns, geometric=FALSE,
> wealth.index=TRUE)
>
>
> Comparing Slides 48 & 57 faber curves do not agree. Slide 48 has a value
> of approx.10.63, Slide 57 has a value of approx.3.77 (wealth+2.77). It
> cannot be both-ways regarding returns from b.strategy, it's either
> geometric or arithmetic chaining?
>
> So hypothetically speaking my boss says to me you have a strategy that
> makes Net profit = 2,770,644, on an initial equity = 1,000,000 (slides
> 30-42), yet you say to me it makes cumulative return of 10.63 (slides
> 47-50), how does that work?

I think you answered your own question.  These are cash returns, without 
compounding.

So applying geometric compounding as though every dollar earned were 
somehow available for instant reinvestment (even though that is clearly 
not what the fixed order sizing of the strategy did) is clearly erroneous.

It seems that analysts need to know when to use geometric versus 
arithmetic chaining, so this is likely a typo in the slide code.  The 
responsibility for understanding that data continues to lie with the reader.

Brian

-- 
Brian G. Peterson
http://braverock.com/brian/
Ph: 773-459-4973
IM: bgpbraverock

_______________________________________________
R-SIG-Finance at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should go.