Message Boards

4980 Views

3 Replies

2 Total Likes

View groups...

Follow this post

Share this post:

GROUPS:

Data Science Mathematics Import and Export Wolfram Language

Import data from a .xlsx file and perform a regression analysis with Fit?

Posted 5 years ago

I am working with a dataset, where I have 51 rows and 4 columns ( I am using SemanticImport to bring in my .xlsx document. ). I want to do some regression analysis, but cannot see how to convert the dataset into coordinates for Fit. How do you get the data into the necessary format? Fit[{education[;; , "SpendPerPupil"], education[;; , "8thMathScore"]}, x, x] is what I tried, but is obviously not right.

POSTED BY: Justin Gatlin

3 Replies

Sort By:

Posted 5 years ago

Thank you. I will look at that Notebook, Ian. I am looking at seeing how different predictors (teacher salary, class size, median household income) correlate with test scores (just out of curiousity). Plotting it in Excel, I saw that teacher salary/median salary is negatively correlated, which piqued my interest. I wanted to bring it into Mathematica to look more deeply. Here are the first few lines of my table. Hope that helps make sense of what I am looking at. Stat......... AverageStartSalary..... MedianIncome SpendPerPupil 8thMathScore Alabama......... $38,477.00............. $44,765.00....... $9,236.00........... 268 Alaska........ $46,785.00................. $73,355.00 ...... $17,510.00......... 277 Arizona ..... $34,068.00 .......... $51,492.00 ....... $7,613.00 .......... 282

POSTED BY: Justin Gatlin

Posted 5 years ago

Hi Justin, I'm not entirely sure what you're trying to do, but the attached notebook (which is based on something I put together while teaching myself how to use the fitting functions in conjunction with datasets) may give you some pointers. One observation on your code is that, assuming your dataset is called "education", I believe you'll need to apply the function Normal to the dataset to transform it into a more conventional list as I don't think Fit operates on pure datasets. This could be easily achieved as follows... Fit[{education[;; , "SpendPerPupil"]//Normal, education[;; , "8thMathScore"]//Normal}, x, x] I hope this helps, Ian Attachments: Fitting data in ...nb

POSTED BY: Ian Williams

Posted 5 years ago

Giving some sample data would likely get you quicker and better help.

POSTED BY: Jim Baldwin

Reply to this discussion

Reply Preview

Attachments

Remove Add a file to this post

Follow this discussion

or Discard

Feedback