# Bug in var() when supplying means as vectors
**URL:** https://discourse.julialang.org/t/bug-in-var-when-supplying-means-as-vectors/36979
**Category:** Statistics
**Created:** [April 3, 2020, 7:34pm UTC](https://discourse.julialang.org/t/bug-in-var-when-supplying-means-as-vectors/36979 "2020-04-03T19:34:09Z")
**Posts on this page:** 5
**Page:** 1
### Author:  [@gabehassler](https://discourse.julialang.org/u/gabehassler)
#### Post date: [April 3, 2020, 7:34pm UTC](https://discourse.julialang.org/t/bug-in-var-when-supplying-means-as-vectors/36979/1 "2020-04-03T19:34:09Z")
I’m trying to standardize (subtract the mean and divide by the standard deviation) of each column in a matrix. In doing so, I think I’v discovered a bug in the var() function when supplying the means.
The following code calculates the variance of the second column of Y as NaN or Inf about 1% of the time.
```julia
using Statistics
n = 1000000
failed = 0
for i = 1:n
Y = [-0.1541210046743307 -1.4703177132816831 0.13467071528166455 -0.09852533308058764 0.877991436467067;
1.0834881229604125 -3.027334287373694 -1.2882941558820766 -1.5294166969571168 -0.7046862823383356;
-0.3527617642771778 -1.9171259707603048 0.38208240655404657 -0.1838590266087666 0.18088977119725;
-0.05385169300191628 -0.5331876236894679 0.13781372390866994 -0.1991690640034743 -0.4252461450819313;
0.847451616756052 0.13212361707950138 -1.3483017839678413 -0.6442240850236565 -0.49114238748086886]
μ = vec(mean(Y, dims=1))
σ2 = var(Y, dims = 1, mean = μ)
if isnan(σ2[2]) || σ2[2] == Inf
global failed += 1
end
end
@show failed / n
```
While the same code without the `vec()` wrapping the means does not result in the same error.
More critically, if I add a column to the array and run the code, I get an EXCEPTION\_ACCESS\_VIOLATION.
The code
```julia
using Statistics
n = 1000000
failed = 0
for i = 1:n
Y = [-0.1541210046743307 -1.4703177132816831 0.13467071528166455 -0.09852533308058764 0.877991436467067 1.0;
1.0834881229604125 -3.027334287373694 -1.2882941558820766 -1.5294166969571168 -0.7046862823383356 2.0;
-0.3527617642771778 -1.9171259707603048 0.38208240655404657 -0.1838590266087666 0.18088977119725 3.0;
-0.05385169300191628 -0.5331876236894679 0.13781372390866994 -0.1991690640034743 -0.4252461450819313 4.0;
0.847451616756052 0.13212361707950138 -1.3483017839678413 -0.6442240850236565 -0.49114238748086886 5.0]
μ = vec(mean(Y, dims=1))
σ2 = var(Y, dims = 1, mean = μ)
if isnan(σ2[2]) || σ2[2] == Inf
global failed += 1
end
end
@show failed / n
```
produces the following error:
> Please submit a bug report with steps to reproduce this fault, and any error messages that follow (in their entirety). Thanks.
> Exception: EXCEPTION\_ACCESS\_VIOLATION at 0x1dac3b1f – getindex at .\array.jl:788 [inlined]
> getindex at .\multidimensional.jl:543 [inlined]
> centralize\_sumabs2! at D:\buildbot\worker\package\_win64\build\usr\share\julia\stdlib\v1.4\Statistics\src\Statistics.jl:247
> in expression starting at C:\Users\gabeh\OneDrive\Documents\Running Scripts\test\_nan.jl:6
> getindex at .\array.jl:788 [inlined]
> getindex at .\multidimensional.jl:543 [inlined]
> centralize\_sumabs2! at D:\buildbot\worker\package\_win64\build\usr\share\julia\stdlib\v1.4\Statistics\src\Statistics.jl:247
> #varm!#10 at D:\buildbot\worker\package\_win64\build\usr\share\julia\stdlib\v1.4\Statistics\src\Statistics.jl:269
> varm!##kw at D:\buildbot\worker\package\_win64\build\usr\share\julia\stdlib\v1.4\Statistics\src\Statistics.jl:265 [inlined]
> \_varm at D:\buildbot\worker\package\_win64\build\usr\share\julia\stdlib\v1.4\Statistics\src\Statistics.jl:299 [inlined]
> #varm#11 at D:\buildbot\worker\package\_win64\build\usr\share\julia\stdlib\v1.4\Statistics\src\Statistics.jl:297 [inlined]
> varm##kw at D:\buildbot\worker\package\_win64\build\usr\share\julia\stdlib\v1.4\Statistics\src\Statistics.jl:297 [inlined]
> \_var at D:\buildbot\worker\package\_win64\build\usr\share\julia\stdlib\v1.4\Statistics\src\Statistics.jl:337
> #var#15 at D:\buildbot\worker\package\_win64\build\usr\share\julia\stdlib\v1.4\Statistics\src\Statistics.jl:335 [inlined]
> var##kw at D:\buildbot\worker\package\_win64\build\usr\share\julia\stdlib\v1.4\Statistics\src\Statistics.jl:335 [inlined]
> top-level scope at C:\Users\gabeh\OneDrive\Documents\Running Scripts\test\_nan.jl:15
> jl\_toplevel\_eval\_flex at /cygdrive/d/buildbot/worker/package\_win64/build/src\toplevel.c:808
> jl\_parse\_eval\_all at /cygdrive/d/buildbot/worker/package\_win64/build/src\ast.c:872
> include\_string at .\loading.jl:1080
> #200 at C:\Users\gabeh.julia\packages\Atom\wlPiw\src\eval.jl:164
> withpath at C:\Users\gabeh.julia\packages\CodeTools\kosGY\src\utils.jl:30
> withpath at C:\Users\gabeh.julia\packages\Atom\wlPiw\src\eval.jl:9
> #199 at C:\Users\gabeh.julia\packages\Atom\wlPiw\src\eval.jl:161 [inlined]
> with\_logstate at .\logging.jl:398
> with\_logger at .\logging.jl:505 [inlined]
> #198 at C:\Users\gabeh.julia\packages\Atom\wlPiw\src\eval.jl:160 [inlined]
> hideprompt at C:\Users\gabeh.julia\packages\Atom\wlPiw\src\repl.jl:140
> macro expansion at C:\Users\gabeh.julia\packages\Media\ItEPc\src\dynamic.jl:24 [inlined]
> evalall at C:\Users\gabeh.julia\packages\Atom\wlPiw\src\eval.jl:150
> jl\_apply at /cygdrive/d/buildbot/worker/package\_win64/build/src\julia.h:1692 [inlined]
> do\_apply at /cygdrive/d/buildbot/worker/package\_win64/build/src\builtins.c:643
> macro expansion at C:\Users\gabeh.julia\packages\Atom\wlPiw\src\eval.jl:39 [inlined]
> #172 at .\task.jl:358
> unknown function (ip: 000000001DA7BF83)
> jl\_apply at /cygdrive/d/buildbot/worker/package\_win64/build/src\julia.h:1692 [inlined]
> start\_task at /cygdrive/d/buildbot/worker/package\_win64/build/src\task.c:687
> Allocations: 309493749 (Pool: 309480613; Big: 13136); GC: 531
Similarly, when I remove the `vec()` wrapper around the mean, there is no error.
I am using Julia 1.4.0 on a Windows 10 machine. I’ve also run this code on Julia 1.3.0 and 1.3.1 with similar results.
---
### Author:  [@nalimilan](https://discourse.julialang.org/u/nalimilan)
#### Post date: [April 3, 2020, 9:40pm UTC](https://discourse.julialang.org/t/bug-in-var-when-supplying-means-as-vectors/36979/2 "2020-04-03T21:40:55Z")
Good catch. Looks like there’s a missing check: the means must have a shape compatible with what `mean(Y, dims=1)` returns, but by calling `vec` you’re reshaping a 1-row matrix into a column vector, so indexing goes out of bounds. I’ll make a pull request to throw an error instead.
---
### Author:  [@nalimilan](https://discourse.julialang.org/u/nalimilan)
#### Post date: [April 5, 2020, 4:07pm UTC](https://discourse.julialang.org/t/bug-in-var-when-supplying-means-as-vectors/36979/3 "2020-04-05T16:07:20Z")
See [Add missing shape checks for the means argument to var[m] and std[m] by nalimilan · Pull Request #32 · JuliaStats/Statistics.jl · GitHub](https://github.com/JuliaLang/Statistics.jl/pull/32).
---
### Author:  [@gabehassler](https://discourse.julialang.org/u/gabehassler)
#### Post date: [April 5, 2020, 4:14pm UTC](https://discourse.julialang.org/t/bug-in-var-when-supplying-means-as-vectors/36979/4 "2020-04-05T16:14:04Z")
Thanks!
---
### Author:  [@laborg](https://discourse.julialang.org/u/laborg)
#### Post date: [April 5, 2020, 4:36pm UTC](https://discourse.julialang.org/t/bug-in-var-when-supplying-means-as-vectors/36979/5 "2020-04-05T16:36:32Z")
You might also make use of the available standardization:
[https://juliastats.org/StatsBase.jl/stable/transformations/](https://juliastats.org/StatsBase.jl/stable/transformations/)