How to calculate the weighted mean?

Hello

I have problems calculating a weighted mean. Let me introduce the background first. I'm implementing a missing value imputation method in the machine learning domain. As a reference I'm using the knnimpute method from Matlab. In the documentation it is written:

knnimpute(Data, k) replaces NaNs in Data with a weighted mean of the k nearest-neighbor columns. The weights are inversely proportional to the distances from the neighboring columns.

Let's say I have a data point x and I'm calculating the nearest neighbours:

[IDX,D] = knnsearch(x,Y,'K',5);

x is 1xD and Y is NxD.

How can I now calculate the weighted mean with weights proportional to the distances?

[Muhammad Usman Saleem]

Tell me what kind of data set you are using ? Geographicb or other? Also tell me what is the dimension of your data whether it is in 1,2, dimension and nan values are in sequence or random?