topic Re: Prediction Interval for Poisson Distribution in Discussions
https://community.jmp.com/t5/Discussions/Prediction-Interval-for-Poisson-Distribution/m-p/187588#M40649
<P>Hi,</P><P> </P><P>Thanks for looking into this question.</P><P>I am using JMP ver 14.</P><P> </P><P>I do know there is a "Prediction Interval" option as you have pointed out.</P><P> </P><P>The reason I am questioning that is because, you can see the Confidence Interval calculated are two totally different value (based on 95%).</P><P> </P><P>In my example:</P><P>Confidence Interval 95% caculated (based on t density function) is 22.6 to 48.1.</P><P>Confidence Interval 95% calculated (based on Poisson density function) is 32.5 to 38.4.</P><P> </P><P>Due to this, it makes me wonder that the Prediction Interval option as you have shown will calculate based on the t density function rather than Poisson.</P><P> </P><P>Hope this brings the question clearly now.</P><P> </P><P>B.r,</P><P>Chris</P>Tue, 19 Mar 2019 02:56:14 GMTChrisLooi2019-03-19T02:56:14ZPrediction Interval for Poisson Distribution
https://community.jmp.com/t5/Discussions/Prediction-Interval-for-Poisson-Distribution/m-p/187183#M40579
<P>Dear JMP Communities,</P><P> </P><P>I have data that are count/space.</P><P> </P><P>In my case, (number of events occured / time). As such, it is naturally I would fit these data using Discrete Fit > Poisson.</P><P> </P><P>Afterwards, JMP produces the fitted Poisson with results of Upper/Lower 95% Confidence Interval per below photo.</P><P> </P><P><span class="lia-inline-image-display-wrapper lia-image-align-left" image-alt="Fitted Poisson.jpg" style="width: 407px;"><img src="https://kvoqx44227.i.lithium.com/t5/image/serverpage/image-id/16344i834AF291644219C6/image-size/large?v=1.0&px=999" title="Fitted Poisson.jpg" alt="Fitted Poisson.jpg" /></span></P><P> </P><P> </P><P> </P><P> </P><P> </P><P> </P><P>My question is : How do I get Upper/Lower 95% <U><STRONG>Prediction</STRONG></U> Interval?</P><P> </P><P>Thanks for the pointer.</P><P> </P><P>B.r,</P><P>Chris</P>Fri, 15 Mar 2019 07:13:06 GMThttps://community.jmp.com/t5/Discussions/Prediction-Interval-for-Poisson-Distribution/m-p/187183#M40579ChrisLooi2019-03-15T07:13:06ZRe: Prediction Interval for Poisson Distribution
https://community.jmp.com/t5/Discussions/Prediction-Interval-for-Poisson-Distribution/m-p/187218#M40582
<P>I don't know which version you are using. Check the inverted red triangle menu next to the variable outline box</P>
<P><span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="image.png" style="width: 386px;"><img src="https://kvoqx44227.i.lithium.com/t5/image/serverpage/image-id/16348i31B1889D1CBC5A54/image-size/large?v=1.0&px=999" title="image.png" alt="image.png" /></span></P>
<P>Here is an example script:</P>
<PRE><CODE class=" language-jsl">Names Default To Here( 1 );
dt = Open( "$SAMPLE_DATA/Quality Control/Airport.jmp" );
obj = Distribution( Column( :Delay ) );
obj << Fit Distribution( Poisson );
obj << Prediction Interval(0.95, 1);</CODE></PRE>Fri, 15 Mar 2019 10:21:44 GMThttps://community.jmp.com/t5/Discussions/Prediction-Interval-for-Poisson-Distribution/m-p/187218#M40582gzmorgan02019-03-15T10:21:44ZRe: Prediction Interval for Poisson Distribution
https://community.jmp.com/t5/Discussions/Prediction-Interval-for-Poisson-Distribution/m-p/187588#M40649
<P>Hi,</P><P> </P><P>Thanks for looking into this question.</P><P>I am using JMP ver 14.</P><P> </P><P>I do know there is a "Prediction Interval" option as you have pointed out.</P><P> </P><P>The reason I am questioning that is because, you can see the Confidence Interval calculated are two totally different value (based on 95%).</P><P> </P><P>In my example:</P><P>Confidence Interval 95% caculated (based on t density function) is 22.6 to 48.1.</P><P>Confidence Interval 95% calculated (based on Poisson density function) is 32.5 to 38.4.</P><P> </P><P>Due to this, it makes me wonder that the Prediction Interval option as you have shown will calculate based on the t density function rather than Poisson.</P><P> </P><P>Hope this brings the question clearly now.</P><P> </P><P>B.r,</P><P>Chris</P>Tue, 19 Mar 2019 02:56:14 GMThttps://community.jmp.com/t5/Discussions/Prediction-Interval-for-Poisson-Distribution/m-p/187588#M40649ChrisLooi2019-03-19T02:56:14ZRe: Prediction Interval for Poisson Distribution
https://community.jmp.com/t5/Discussions/Prediction-Interval-for-Poisson-Distribution/m-p/187647#M40652
<P>You are correct - the prediction interval for the lambda is not available in the Distribution platform. Equivalently, I fit an intercept-only Poisson loglinear regression model with the GLM platform to obtain the same estimates, but it does not provide prediction intervals either.</P>Tue, 19 Mar 2019 10:00:22 GMThttps://community.jmp.com/t5/Discussions/Prediction-Interval-for-Poisson-Distribution/m-p/187647#M40652markbailey2019-03-19T10:00:22Z