# Beta distribution

#### menahg

##### New Member
Hi,
I want to have a distribution similar to U-shape beta distribution as the following:

I want to make the distribution as :

Any idea?

Thanks

Last edited:

#### Dason

What do you mean by "left up the curve"?

#### menahg

##### New Member
What do you mean by "left up the curve"?
I mean to move the curve to the top, I want to increase the probability of the values between 0 and 1

#### Dason

All of the values are between 0 and 1. I'm just not understanding you. Can you make a sketch of what you're talking about? Or try phrasing it differently?

#### menahg

##### New Member
All of the values are between 0 and 1. I'm just not understanding you. Can you make a sketch of what you're talking about? Or try phrasing it differently?
Thanks @Dason, I added an image.

#### hlsmith

##### Less is more. Stay pure. Stay poor.
Not following @Dason comment (but he is the best), but isnt that just a ~Beta (0.5, 0.5)?

#### hlsmith

##### Less is more. Stay pure. Stay poor.
Oh, I get the request now.

@menahg can you explain the context you are working in or what this will be used for?

#### menahg

##### New Member
Oh, I get the request now.

@menahg can you explain the context you are working in or what this will be used for?
@hlsmith I intend to use it for a simulation, to model a decision for sharing an opinion among agents. It is a Beta distribution, but I want to increase the probability in the middle a little.

Last edited:

#### hlsmith

##### Less is more. Stay pure. Stay poor.
Thanks for sharing, did you find the right a and b values to achieve that?

#### menahg

##### New Member
Thanks for sharing, did you find the right a and b values to achieve that?
I used 0.5 for both a and b. Thanks

#### hlsmith

##### Less is more. Stay pure. Stay poor.
Probably not an option, but in Bayesian modeling I keep seeing the "horse shoe" prior which has a similar shape.

#### menahg

##### New Member
Probably not an option, but in Bayesian modeling I keep seeing the "horse shoe" prior which has a similar shape.
I will look into it. Thank you @hlsmith!