# Right angled Triangular distribution

I have an empirical probability distribution which looks somewhat like a right angled triangle. I am trying to model a process which relys on a semi accurate model of this distribution.

There isn't really a distribution that springs to mind instantly about this particular density plot.

So I ask, what distribution, if any do you think, best fits this particular PDF? By a right angled distribution i'm thinking something like this: • What are the data? Is this something that is supposed to be non-negative? (You do seem to have 1 negative value.) – gung - Reinstate Monica Feb 15 '16 at 0:43
• It is possible to have negative data values, however this is rare and in my opinion generally unimportant to the model – aeongrail Feb 15 '16 at 0:47
• I'm trying to predict a time something will occur, the data values are the predicted time/the actually time the thing occured – aeongrail Feb 15 '16 at 0:48
• What does that mean exactly? Are you predicting the hour of the day, the length of time you have to wait from a starting point till something occurs, etc? What are the data? Are these response / Y values, or residuals from a model? What model, what type of residuals? We will need to know more before anything can be said. – gung - Reinstate Monica Feb 15 '16 at 0:50
• There is such a distribution as you have drawn; it's a special case of the more general triangular distribution. For this special case, once you define the location of the mode and the right endpoint, the entire distribution is determined. – Glen_b Feb 15 '16 at 1:20

## 1 Answer I've found one possible option to fit the data, an exponential PDF is one possible option

• It's not exponential because you can see at least one of the data values must have been negative. – Glen_b Oct 23 '16 at 7:25