Mathcad Prime 3,4,5

[Joeboy]

I wiould like to see a probability based x axes posible

I would also like to see useful error messages, particularly as they pertain to solve blocks. In general I feel Mathcad's error messages leave a lot to be desired.

[mzeftel]

Dixie,

We are planning to improve error message, especially in solve blocks for the next release. What do you mean by probability based x axes? Can you provide an example?

Thanks,

Mona

[Joeboy]

Mona, I am talking about what is commonly called "probability paper", normal distribution probability paper has an x axis that runs from a very small number, say .001 to 99.99 or so. Data that is normally distributed will plot as a straight line on such paper. There is also log probability paper, where the y axis has a log scale. There are also papers for other statistical distributions such as the Gumbel distribution. All are used alot in hydrology. I suspect the MIT bookstore has it in stock.

[Joeboy]

What would really be nice if there was a pull down menu where on e could chose the type of x and y scale desired. This could allow for a variety of probability distributions to be available. The usefulness of any probability scale is that data that fits that probability distribution will plot as a straight line on that scale, used a lot in hydrolgy assess various types of flow data.

[Joeboy]

Hey Mona, while I'm at it what can I do with the points I'm getting here on PTC ? Plane tickets ?, Cruises ?, trips to Europe ?

[mzeftel]

I'm waiting for my prize too.

Dan Marotta manages the community, so you'll have to ask him.

I think you can do some of what you want in Mathcad Prime 2.0. If you click on either the x-expression or the tick mark values for the x-axis you can choose Plot tab > Axes > Logarithmic Scaling. The same is true for the y-axes.

And I'll add a feature suggestion for the remainder of your requests.

Thanks,

Mona

[VladimirN]

Mona Zeftel wrote:

...Can you provide an example?...

Like this:

P.S. More here: http://www.weibull.com/GPaper/

[Joeboy]

Attached is a sample of log probability paper. I do not prefer the standard normal variate on the y axis