The download includes the current version of 'SRS1 Cubic Spline for Excel' (version 2). This is probably the version that you would want to use, but it also includes an older version (Version 1.xx) and a fully functional trial version of Data Curve Fit Creator Add-in.

The new version (Version 2) of the software requires Excel 2007 or later. It also runs faster, is more robust, and will have fewer issues with security software since it is a compiled .net add-in. If you have Excel 2003 or earlier, then you would need to use the earlier version. Also, the earlier version is written in VBA, so you can see the code within that version. One other benefit is that if you add the earlier version's VBA code to a workbook, and then give the workbook to someone on another machine, then they don't need to install anything to get the function to work.

Version 2 has two functions: 'cubic_spline' and 'linear_interp'. These functions are listed under the 'SRS1Spline.Functions' function category in the 'Insert Function' dialog box of Excel. The older version (Version 1.xx) only has the 'cubic_spline' function, and it is listed under the 'User Defined' function category of the 'Insert Function' dialog. If 'SRS1 Cubic Spline for Excel' is installed, then if you type '=cubic_spline(' in a cell in Excel, then the Excel function wizard will list the arguments that you need to provide for the function.

A possible cause for this is that the Excel security settings are blocking macros from running on the workbook. Adjust the security settings to at least prompt and ask you whether you want to allow macros and/or VBA code to run. Do NOT just turn off all security settings, as this could potentially leave Excel unprotected from security threats.

When you see '#NAME?' in an Excel cell, then it means that Excel does not recognize the function that you are trying to use. If you see this then you should first make sure that you are spelling the name of the 'SRS1 Cubic Spline for Excel' function correctly. The correct names are 'cubic_spline' and 'linear_interp'. If you are sure that you have the name correct, then that means that add-in is not installed or is not enabled. You can try installing the software again. If you are using the old version of the software (version 1.xx) then make sure that your security settings in Excel are set to allow the custom function to run.

Yes, but 'Data Curve Fit Creator Add-in' has its own cubic spline and linear interpolation functions, as well as many more advanced functions and options. Note also that there is no function name conflict when you have both installed. 'SRS1 Cubic Spline for Excel' uses underscores in its function names ('cubic_spline', 'linear_interp'), but 'Data Curve Fit Creator Add-in' does not use underscores (e.g. 'CubicSpline', 'LinearInterp').

A cubic spline interpolates a smooth line through a given set of data points. Unlike a polynomial or other curve fits, a spline is forced to directly pass through all points in the data set. There are a few other constraints that are used to generate the smooth curve. Interpolating between two points, if we know the second derivative tat the first and second points, then the second derivative of the interpolated curve will vary linearly from the first to the second point. Also, the first derivative of the curve needs to be continuous across the curve.

As mentioned before, the cubic spline is forced to go through all data points. If there is noise in your data, then the curve can become erratic as it tries to smoothly pass through each point. Some things that you can do to improve the behavior of the curve include removing bad data points or smoothing the data before creating the spline.

Your source data input values (X values) need to be in ascending order, or else the function will not work correctly.

This version of the cubic spline does not allow you to directly specify the slope at a data point. There is a kluge that you can try in order to achieve this, though. You can try to force the slope at a data point to be what you want by adding a fake data point right next to it at the desired slope. For example, to set the slope at point [x=1, y=6] to be zero, you could add a new data point [x=0.999, y=6], and then maybe one more at [x=1.001, y=6]. Again, this is definitely a kluge, but in some cases it could be useful. Note also that Data Curve Fit Creator Add-in includes a 'FlexSpline' function that lets you specify the slope at any point.

We recommend the book 'Numerical Recipes: The Art of Scientific Computing, Third Edition (2007)', published by Cambridge University Press'. The code in SRS1 Cubic Spline for Excel is based on the cubic spline in that book.