Wrote an article - score it - GoFuckYourself.com

fluffygrrl · 05-21-2008, 08:27 AM

Is my sponsor fucking me over ?

Tom_PM · 05-21-2008, 08:28 AM

Why dont you post it? I dont wanna click linkies.

fluffygrrl · 05-21-2008, 08:45 AM

Quote:

Originally Posted by PR_Tom

Why dont you post it? I dont wanna click linkies.

It's friggin long tho, like 2k words.

ADL Colin · 05-21-2008, 08:47 AM

Funny cuz i wrote an article with similar reference to Poisson distribution
http://www.buildinganempire.com/poisson2.html

fluffygrrl · 05-21-2008, 09:00 AM

Quote:

Originally Posted by ADL Colin

Funny cuz i wrote an article with similar reference to Poisson distribution
http://www.buildinganempire.com/poisson2.html

Well, actually, you seem to missunderstand the math.

Quote:

The interesting point of all this is that the error depends only on the number of signups and not the number of visitors

This is actually false. lambda varies with the number of visitors, k varies with the number of sign-ups, the probability depends on both.

Obviously, things fall apart from there on,

Quote:

Most people think, ok, I'll send a certain number of visitors to a sponsor and see how many signups I get. What you should be doing instead is saying ok, I'll send a certain number of signups to a sponsor to test them.

Which is patent nonsense.

Affiliates have to make the following business decission : I've sent X visitors to a sponsor with 0 conversions so far, how long do I give it before I pull the plug ?

With the connoted question of, how likely is this to be a mere statistical artefact, and how likely is it to be a bad fit for conversions ?

So, it's nice that you've tried, but I think you should try re-reading on the subject.

PS. Do the exercises too, it will help.

fluffygrrl · 05-21-2008, 10:51 AM

Priceless, your article, it turns out. Thanks for linking.

Quote:

Sending enough hits to test a sponsor There's a concept in statistics called standard deviation. The idea is that 68% of the time a quantity will be within one standard deviation of the average. 95% of the time one expects to find the quantity within two standard deviations of the average. When a scientist measures some quantity they quote it with an error measurement which is one standard deviation. No measurement is certain. I will just refer to the standard deviation as the error, but you will know that technically I mean the standard deviation. I've also left a few technical details out such as how confident we are in our measurements because well, I'm only going to take this so far.

Quote:

In the Poisson Distribution the error is simply the square root of the average.

I always thought standard deviation is somehow related to the statistical variance, for whatever reason. But I guess we might as well re-christen variance and call it the average. Altho I always thought it's a mean not an average. But I guess we could re-christen means to be averages too.

Bloody hell I haven't laughed this much in months.

68%.

ADL Colin · 05-21-2008, 12:46 PM

Quote:

Originally Posted by fluffygrrl

Priceless, your article, it turns out. Thanks for linking.

I always thought standard deviation is somehow related to the statistical variance, for whatever reason. But I guess we might as well re-christen variance and call it the average. Altho I always thought it's a mean not an average. But I guess we could re-christen means to be averages too.

Bloody hell I haven't laughed this much in months.

68%.

As this page says - "Under the conditions where the Poisson distribution is applicable, the standard deviation may be approximated by the square root of the mean." Is it the word "approximately" you felt is missing? Or maybe you would prefer "mean" over "average" but I think most people would understand average better.

ADL Colin · 05-21-2008, 12:58 PM

Quote:

Originally Posted by fluffygrrl

Affiliates have to make the following business decission : I've sent X visitors to a sponsor with 0 conversions so far, how long do I give it before I pull the plug ?

Of course you have to make a business decision which is exactly why i stated that "No mathematical distribution can make a decision for you, but they can give you a very good idea of when enough is enough. The question then becomes, what probability level is enough for you?" with further elaboration both before and after.

Phil · 05-21-2008, 01:06 PM

solid F as in "fucking fat idiot". congrats

fluffygrrl · 05-21-2008, 02:18 PM

Quote:

Originally Posted by ADL Colin

As this page says - "Under the conditions where the Poisson distribution is applicable, the standard deviation may be approximated by the square root of the mean." Is it the word "approximately" you felt is missing? Or maybe you would prefer "mean" over "average" but I think most people would understand average better.

A mean is not the same as an average Colin.

Other than that, you're confusing lambda with k. in my formula, or e and m in yours, or a and x in the formula on this latest page you quote.

Let's get them clear :

lambda, e and a equal the EXPECTED number of occurences. Thus, for 1:100 conversions, and 900 clicks, lamba, e and a equal 9.

k, m and x equal the ACTUAL number of occurences. Thus, that'd be the number of actual sign-ups.

Now, the standard deviation, which is the root of the the variance, can be approximated as the square root of the mean of lamba, e or a, depending on which formula you use.

Which makes it the square root of the number of expected signups, not actual signups, as contrary to

Quote:

In the Poisson Distribution the error is simply the square root of the average. In other words, the error is the square root of the number of signups.

Much more interesting are the points you don't answer to, namely

1. The false statement that signup probability is independent of the number of visitors,
2. The false statement that one should not think "I'll send X visitors" but instead "I'll send X sign-ups"

The entire point of using this math is to establish how many visitors to send, and under which feedback conditions.

And while you claim math can not make a decission for you, that is only because you don't quite know how to use it.

Properly applied, this bit of math will tell you, after having sent x visitors, if it's worth trying to send k more, or not.

fluffygrrl · 05-21-2008, 02:19 PM

Quote:

Originally Posted by CAMOKAT

solid F as in "fucking fat idiot". congrats

You sure have a hard-on for me dontcha.

2012 · 05-21-2008, 02:36 PM

Quote:

Originally Posted by ADL Colin

Funny cuz i wrote an article with similar reference to Poisson distribution
http://www.buildinganempire.com/poisson2.html

great article thanks

fluffygrrl · 05-21-2008, 03:00 PM

Quote:

Originally Posted by fartfly

great article thanks

Whoa I can't believe somebody posted something nice in a thread of mine ?

Isn't this like in the prophecy of the apocaplypse and all ?!

BradM · 05-21-2008, 03:03 PM

Worthless as usual. Every word you type just sinks into an abyss.

fluffygrrl · 05-21-2008, 04:29 PM

Quote:

Originally Posted by BradM

Worthless as usual. Every word you type just sinks into an abyss.

Give BradM a stick, he'll beat himself silly with it.

Phil · 05-21-2008, 07:03 PM

bump for fat ass retard and more useless shit.

ADL Colin · 05-22-2008, 08:05 AM

Quote:

Originally Posted by fartfly

great article thanks

You're most welcome :-)

View Poll Results: I'd give you an
F	12	80.00%
D	1	6.67%
C	1	6.67%
B	1	6.67%
A	0	0%
Voters: 15. You may not vote on this poll

05-21-2008, 08:27 AM	#1
fluffygrrl So Fucking Banned Join Date: May 2006 Posts: 2,187	Wrote an article - score it Is my sponsor fucking me over ?

05-21-2008, 08:28 AM	#2
Tom_PM Porn Meister Industry Role: Join Date: Feb 2005 Posts: 16,443	Why dont you post it? I dont wanna click linkies. __________________ 43-922-863 Shut up and play your guitar.

05-21-2008, 08:47 AM	#4
ADL Colin Too lazy to set a custom title Join Date: Feb 2001 Location: Tube Titans, USA Posts: 11,929	Funny cuz i wrote an article with similar reference to Poisson distribution http://www.buildinganempire.com/poisson2.html __________________ Adult Date Link - $50 PPS starting NOW! -- good and JUICY! skype = "adultdatelink"

05-21-2008, 01:06 PM	#9
Phil Confirmed User Join Date: Jan 2004 Posts: 7,659	solid F as in "fucking fat idiot". congrats __________________ Ask Phil

05-21-2008, 03:03 PM	#14
BradM Confirmed User Join Date: Dec 2003 Location: 1123,6536,5231 Posts: 3,397	Worthless as usual. Every word you type just sinks into an abyss.

05-21-2008, 07:03 PM	#16
Phil Confirmed User Join Date: Jan 2004 Posts: 7,659	bump for fat ass retard and more useless shit. __________________ Ask Phil