Alex Arslan

It appears to be because of this: https://github.com/JuliaLang/julia/blob/master/stdlib/Statistics/src/Statistics.jl#L134 ``` julia> x = [(i == j == 3) ? missing : i * j for i in 1:3, j in 1:4]...

Going further, we get that result from `reducedim_init` because of this line: https://github.com/JuliaLang/julia/blob/master/base/reducedim.jl#L117, since `zero(Union{Missing,Int}) === 0`.

`sum` handles this just fine, so I think we should change `mean` to go though more of the `sum` machinery.

While I agree that the user experience isn't optimal, note that this isn't specific to `mean`: `sum` exhibits identical behaviors for these examples. It comes down to whether the correct...

The difference here makes sense. The median is by definition either a data value (if your list is of odd length) or the arithmetic mean of the middle two values...

Throwing an error from `mean`, `var`, etc. seems reasonable to me.

Unicode should not be a required part of the API, so the keyword argument should be `mean` instead of `µ`, but otherwise I support that idea.

That's a really cool idea! My only concern is with the proposed API, since for similar functions like `mean`, `mean(f, x)` means `mean(map(f, x))`. It would be kind of weird...

This package seems no longer actively maintained. I had a go at getting it working on 0.5 a while back (see #18) but didn't have a lot of luck. I...

Unfortunately there's no working branch for this package. For SVM, you can try `Pkg.checkout("SVM", "master")`.