I have a simple set of X-Y data as shown below. When I chart the data using a scatter plot and Excel 2007, it's apparent by eye that the points fall on a good straight line. If I use: Chart Tools/Layout/Trendline and select a linear fit, a line is fit to the data points with an R^2 value of 0.9955 and an equation:
y = -4E+07X - 1.9692. This all makes sense and confirms what the eye sees on the chart.
If I use the LINEST function on the same data set, it returns the nonsensical values of slope and Y-intercept shown below. Note: I have not forced a zero intercept, the X- and Y-ranges are not overlapping and the number of data rows exceeds the number of data columns, so none of the known reasons for LINEST giving incorrect results is present.
Any explanations?
y = -4E+07X - 1.9692. This all makes sense and confirms what the eye sees on the chart.
If I use the LINEST function on the same data set, it returns the nonsensical values of slope and Y-intercept shown below. Note: I have not forced a zero intercept, the X- and Y-ranges are not overlapping and the number of data rows exceeds the number of data columns, so none of the known reasons for LINEST giving incorrect results is present.
Any explanations?
Excel Workbook | |||||||
---|---|---|---|---|---|---|---|
A | B | C | D | E | |||
1 | from LINEST | ||||||
2 | X | Y | Slope | Y-Intercept | |||
3 | 7.3E-07 | -28.9536 | 0 | -32.5707215 | |||
4 | 7.6E-07 | -30.5338 | |||||
5 | 8.2E-07 | -32.2276 | |||||
6 | 8.7E-07 | -34.3473 | |||||
7 | 9.36E-07 | -36.7913 | |||||
Sheet1 |