# Help understanding the use of Linest function

1. ## Help understanding the use of Linest function

I understand this is a bit more of a math question than an excel question, so if it's beyond the scope of this forum that's okay. I'm trying to model multi variable linear regression. I have all the data I need but I don't really understand what the data means. I know it gives me beta/slope values for each variable, but where I get lost is how the beta values are affected by co linearity. I know that somehow each beta value represents how much that variable adds relative to the other variables but I'm not sure what to do with that information, or how you're able to determine how much information is added. Lastly I don't really know what to do with this information. I'm supposed to create a model that determines the highest chance of achieving a y value of 1 (indicating a credit card default in this context), but I'm not sure how to use the linest results to do that.

I've included a screenshot of the data with the linest function applied. For some reason the attachment button isn't working so a screenshot will have to do until I can get it to work. Sorry!

linest screenshot.JPG

2. ## Re: Help understanding the use of Linest function

I'm not sure I understand your question, or if we even have a statistician who would adequately understand the issues around collinearity.

A quick internet search found:
Wikipedia: https://en.wikipedia.org/wiki/Multicollinearity
a Penn State lesson outline: https://onlinecourses.science.psu.edu/stat501/node/345
I find a bias against Excel among many statisticians, but this guy specifically targets the Excel audience: http://www.real-statistics.com/multi.../collinearity/

There are many other internet resources for studying collinearity. Your question sounds more like you want to understand the statistics behind collinearity. I expect that once you review the statistics, programming the collinearity tests into Excel will be much easier.

As for the question of using the results, the output of a multivariate regression should be an equation something like y=m1x1+m2x2+m3x3+... Once you have the m's from the LINEST() function, you should be able to put your values for the x's in and calculate y, or whatever you need to do.

Does any of that help?

There are currently 1 users browsing this thread. (0 members and 1 guests)