We want to hear your ideas for improving JMP software.

Search: Please search for an existing idea first before submitting a new idea.

Submit: Post your new idea using the Suggest an Idea button. Please submit one actionable idea per post rather than a single post with multiple ideas.

Kudo & Comment Kudo ideas you like, and comment to add to an idea.

Subscribe: Follow the status of ideas you like. Refer to status definitions to understand where an idea is in its lifecycle. (You are automatically subscribed to ideas you've submitted or commented on.)

We consider several factors when looking for what ideas to add to JMP. This includes what will have the greatest benefit to our customers based on scope, needs and current resources. Product ideas help us decide what features to work on next. Additionally, we often look to ideas for inspiration on how to add value to developments already in our pipeline or enhancements to new or existing features.

The set of calculators available in Help -> Sample Index does not include a calculator for hypothesis testing of 1 variance or std deviation. Can this be added?

Hi @MShroff, most of the calculators in the Sample Index are also natively available from JMP Platforms if you're working with raw data. The hypothesis test for one standard deviation is available from the Distribution platform. Does this meet your needs?

In many cases, I don't have raw data and only have summary statistics, so having a calculator would help.

On a related note, I saw that the calculator for hypothesis testing of 2 variances uses the F ratio. Why is this the case rather than a Chi-Sq test? Can an option be added to allow the user to select between F ratio and Chi-Sq?

I can see how a hypothesis test for 1 variance could be useful from summary data, and we can investigate with the developer of these applets the feasibility of adding this.

Regarding the test for equality of two sample variances, I am not familiar with a two-sample approach based on a Chi-Square that is robust to distributional violations (which is why I'd choose something other than the F test). Bartlett's chi-square is all I can think of, but it is not so robust to violations of the assumption of normality. My go-to if an assumption of normality is questionable is Levene's Test. Would you provide a source for what you're interested in?

I was informed by Laura Archer in JMP Tech Support that the F-ratio presented in the 2-variance test is the ratio of the chi-sq metric for each variance. If so, it would be helpful to additionally publish these values.

Thank you for clarifying, @MShroff! I see why you're asking about a Chi-Square now. To clarify, sample estimates of a population variance are themselves distributed as a chi-square. An F is defined as a ratio of two chi-squares, which is how we're able to form a test for the homogeneity of variance by using that ratio as a test statistic compared against an F with n1-1 numerator and n2-1 denominator degrees of freedom.

You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.