Stacked Confidence Intervals

[Binator]

Hi.

I am trying to create a simple means of displaying a set n ( whose values can set by a slider) confidence intervals for a population proportion based on samples of size 100 taken from a any distribution.

The attached file uses sampling from the Binomial Distribution and shows my progress so far in creating the random intervals - not a lot of progress, I must confess. I cannot see how to use cell references to construct a function in the graping window of CAS CX of the form

                          

This may be the totally wrong way of going about this. So I welcome suggestions on how to proceed. 

[Dennis Donovan]

I have attached a file that shows a way I have done similar things:

I use lists and a scatterplot on a graph page.

You want to graph the points in order (a1,y1), (b1,y1), then a null point, then (a2,y2), (b2,y2) then a null point, etc.  Where (a1, b1) is interval #1, (a2, b2) is interval #2, etc. and y1, y2, etc. are arbitrary y values to display the intervals horizontally.

The Attributes for the scatterplot should be changed to Thin points and Connected.

The reason you need the null point is so that you will have disjoint line segments on this connected scatter plot.

It is probably easiest to create the lists this way by using a program.  Then run the program on a notes page so it is constantly running and updating any changes that you make to the samples, n or the confidence level.

---

From: Binator <padica...@gmail.com>
To: tinspire <tins...@googlegroups.com>
Sent: Monday, November 30, 2015 5:36 PM
Subject: [tinspire] Stacked Confidence Intervals

--
--
To post to this group, send email to tins...@googlegroups.com
To unsubscribe send email to tinspire+u...@googlegroups.com
For more options, visit this group at
http://groups.google.com.au/group/tinspire?hl=en-GB?hl=en-GB
The tns documents shared by group members are archived at
https://sites.google.com/site/tinspiregroup/classroom-news/welcome-abouttime

---
You received this message because you are subscribed to the Google Groups "tinspire" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tinspire+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Dennis Donovan]

Try this out.

I believe this is what you want.  I added the ability to re-take the samples and to change the confidence level.  I think 50 intervals is as many as you would want to display on the screen, given its size.

I left everything wide open so you could see what I did.  Check out the formula row on the L&S page and I calculated the Z* value on the Notes page.  The program buildlst(a, b) runs on the notes page as well and will dynamically update the xc and yc lists as things change.

I'm glad you asked about this, I've been meaning to do this for some time.

Dennis Donovan

T3 Regional Instructor

Abington, MA

---

From: Binator <padica...@gmail.com>
To: tinspire <tins...@googlegroups.com>
Sent: Monday, November 30, 2015 5:36 PM
Subject: [tinspire] Stacked Confidence Intervals

[Binator]

Greetings Dennis.

You are correct. This is exactly what I wanted to do.

Thank you for your labours.

Sincerely,

James

[Dennis Donovan]

James,

Here is an updated version that shows the intervals that missed the true proportion in red.  You can also show that by increasing the confidence level that you can turn some intervals from red to blue.

Dennis

---

From: Binator <padica...@gmail.com>
To: tinspire <tins...@googlegroups.com>
Sent: Monday, November 30, 2015 5:36 PM
Subject: [tinspire] Stacked Confidence Intervals