Know Your Data with Continuous Probability Distribution

@Swetha_Kannan

Any plans to add the lognormal distribution to your workflow. Its very commonly used to represent severity distributions of insurance losses.

Thanks.

The csv files aren’t in the data folder.

did they leave without saying goodbye?

@rfeigel which workflow are you referring to? I searched the Hub briefly but didn’t find a match. If you have a link handy I can check.

The files aren’t present in the workflow data folder, but rather in separate folder that’s part of the broader space on the KNIME Hub:

I downloaded all of the workflows here as a single KNAR package and they executed OK for me. The easiest way to do that is to click the cloud icon (note that you must be logged into the Hub):

1 Like

As far as your initial question, not sure if there were plans to update this further to add more distributions. Maybe @roberto_cadili can comment.

The Python fitter actually fits ~80 distributions. I didn’t realize when I asked the question. As to the data location - its pretty confusing since the csv reader points to the empty data folder.

Hi @rfeigel thanks for your questions and feedback :slight_smile:. That’s correct, the Python fitter library at the basis of the component fits approx. 80 distributions and if I am not mistaken the lognormal distribution is one of them.

Next week, part 2 of “Know Your Data with Continuous Probability Distribution” will be published and there we describe the implementation with KNIME.

As for accessing the data, we decided to place it in the Data folder in the current workflow area because we are conceiving the space as a compact learning experience (in the same way we do for the workflows of our courses) where you can learn from descriptive statistics to probability distributions and more by downloading one workflow group only. The next workflow on which we are currently working on is hypothesis testing.

I hope it helps. Happy KNIMEing! :slight_smile:

Roberto

1 Like

@roberto_cadili Thanks for the reply. I’m still confused about the data location. I downloaded the workflow from the recent blog and couldn’t find the data anywhere.

Hi @rfeigel, you find the data in the Data folder in the Statistics with KNIME space. Read once again what @ScottF (thanks!) wrote in a previous comment: Know Your Data with Continuous Probability Distribution - #6 by ScottF. :wink:

If you download the entire space using the cloud icon that Scott marked in red, you will be able to execute the workflow without problems. If you are confused because you cannot see that cloud icon, the answer is very simple: you need to be logged in in the KNIME Community Hub :slight_smile:.

Hope it helps, happy KNIMEing! :slight_smile:

Roberto

1 Like

Hi @rfeigel here is part 2, where we explore the KNIME implementation: https://www.knime.com/blog/identify-continuous-probability-distribution. Enjoy the blog post! :slight_smile:

1 Like

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.