Calculation for logarithmic slope

[lmayo]

I am trying to modify a spreadsheet so that we can plug the slope off of a log regression equation into a formula, however, I cannot figure out how excel is calculating slope. The given formula is y=-0.098ln(x)-0.3401. My max and min data points are (69,0.375758) and (585,0.157576) and the data is a pretty decent fit along that line (R2=.82). Any ideas on how they got from one to the other?

I need to get the slope into the form of ln/sec in order to plug it into my spreadsheet.

 

[shg]

The slope of y=-0.098ln(x)-0.3401 is -0.098



EDIT, Sorry, I get it. Hold on a sec ...

EDIT2: =LINEST(y, LN(x))

 

[lmayo]

Yes, but I am wondering how you would actually calculate that using a regular calculator. What is the equation that they are using?

 

[shg]

It's a least-squares regression of y onto LN(x). You would solve it via matrix inversion, which would be a PITA on a calculator -- at least on the last calculator I used before I stopped using them.

EDIT: Actually, for a simple linear regression in one variable, it's simpler than that ...

EDIT2: See Help for the Slope function. Substitute LN(x) for x in the equation.

 

[tusharm]

See
Trendline Coefficients and Regression Analysis
http://www.tushar-mehta.com/publish_train/data_analysis/16.htm
particularly the section 'Regression analysis with other trend functions'

Yes, but I am wondering how you would actually calculate that using a regular calculator. What is the equation that they are using?