Estimating missing values with kernel regression

Scene 1:

Hello, and welcome. In this demo, we will use the non-parametric NumXL kernel regression function to calculate values for observations with empty or missing values.

First, add the X values we wish to calculate in the adjacent column.

Next, select the top cell in the column adjacent to the X-values and type NxKREG in the formula bar. Then, press the FX button on the left of the Excel function.

The Excel function arguments Window pops up.

For the X-argument, select the cells in column A.

For the Y-argument, select the cells in volume B.

For the polynomial order argument, type two(2) for quadratic.

For the kernel weight function argument, type six(6) for the Gaussian kernel.

For the kernel bandwidth argument, type one or leave it empty.

For the optimization switch argument, type one(1) or true to compute an optimal kernel bandwidth value.

Next, select the cells with the desired x values in column D for the Target argument.

Finally, enter zero(0) or leave it blank for the return type argument to return the fitted (AKA) forecast values.

Now, press the OK button.

Here you go. The function returns all fitted values using the kernel weights function with optimal bandwidth.

Great. Check out the fitted values and original data points on one graph.

That is it for now. Thank you for watching!