Probability Density Function needed

May 2010
7
0
I am looking for a continuous probability density function defined in the interval [0, 1] with the following properties:

  1. If the mode is 0.5 then the mean is also 0.5
  2. If the mode is < 0.5 then: mean < mode
  3. If the mode is > 0.5 then: mean > mode
  4. The first derivative of the probability density function must be a continuous function.
I also wish to be able to have a shape parameter to control the distance between the mean and the mode

It would be even better if I could control the position of the mean relative to the mode via the parameter.

Does anyone knows such a function?

Thanks.
 
Last edited:

CaptainBlack

MHF Hall of Fame
Nov 2005
14,972
5,271
someplace
I am looking for a continuous probability density function defined in the interval [0, 1] with the following properties:

  1. If the mode is 0.5 then the mean is also 0.5
  2. If the mode is < 0.5 then: mean < mode
  3. If the mode is > 0.5 then: mean > mode
  4. The first derivative of the probability density function must be a continuous function.
I also wish to be able to have a shape parameter to control the distance between the mean and the mode

It would be even better if I could control the position of the mean relative to the mode via the parameter.

Does anyone knows such a function?

Thanks.
I would look at the Beta distribution if I were you.

CB
 
May 2010
7
0
I would look at the Beta distribution if I were you.

CB
Thanks. I have been working with Beta but I could not find a set of rules or an algorithm for the α and β parameters that would fulfill the requirements.
Could you give it a try?
Best regards
Paulo
 

matheagle

MHF Hall of Honor
Feb 2009
2,763
1,146
The mean is \(\displaystyle {\alpha\over \alpha+\beta}\) what do you have as the mode?
 
  • Like
Reactions: pandre
May 2010
7
0
The mean is \(\displaystyle {\alpha\over \alpha+\beta}\) what do you have as the mode?
I am using the mode as
\(\displaystyle
{\alpha-1\over\alpha+\beta-2}
\)

I fail to see how the Beta Probability Density Function can comply to the requirements I have. Note that the requirements are such that the mean is always in the shorter tail, and on Beta PDF the mean is, as far I know, always in the longer tail.

Regards,

Paulo
 
Last edited:

matheagle

MHF Hall of Honor
Feb 2009
2,763
1,146
Ok, I just checked that.
I get that as the mode too.
 
Last edited:

matheagle

MHF Hall of Honor
Feb 2009
2,763
1,146
yup the Captain's right.

If \(\displaystyle {a\over a+b}=1/2\) then \(\displaystyle a=b\)

So \(\displaystyle {a-1\over 2a-2}=1/2\).

And if \(\displaystyle {a-1\over a+b-2}<1/2\) then \(\displaystyle a<b\)

so, adding a to both sides we have \(\displaystyle 2a< a+b\) or \(\displaystyle {a\over a+b}<1/2\)
 
May 2010
7
0
Matheagle,
thanks for the effort, but I knew this and it does not fulfill the requirements.
I know that, for the Beta PDF, Mode < 0.5, implies Mean < 0.5. However this also implies that Mean > Mode and what I am looking for is that, in this case, Mean < Mode. And the Beta PDF cannot, as far as I see, fulfill the requirements.
 

CaptainBlack

MHF Hall of Fame
Nov 2005
14,972
5,271
someplace
Matheagle,
thanks for the effort, but I knew this and it does not fulfill the requirements.
I know that, for the Beta PDF, Mode < 0.5, implies Mean < 0.5. However this also implies that Mean > Mode and what I am looking for is that, in this case, Mean < Mode. And the Beta PDF cannot, as far as I see, fulfill the requirements.
\(\displaystyle
\alpha<\beta \Rightarrow mode<0.5
\)

\(\displaystyle
\alpha+\beta>2 \Rightarrow mode<mean
\)

and a similar construction will give the other case/s

CB
 
  • Like
Reactions: pandre
May 2010
7
0
Captain,

That is the inverse of what I am looking for!!

I require that in this case Mode > Mean

Another way of saying it is that for Mode < 0.5 I require that the Mean is at the left of the Mode.