# Using Survey/Inverse Probability Weights in Regression
**URL:** https://discourse.julialang.org/t/using-survey-inverse-probability-weights-in-regression/10444
**Category:** Statistics
**Created:** [April 19, 2018, 5:57pm UTC](https://discourse.julialang.org/t/using-survey-inverse-probability-weights-in-regression/10444 "2018-04-19T17:57:22Z")
**Posts on this page:** 4
**Page:** 1
### Author:  [@jhartma](https://discourse.julialang.org/u/jhartma)
#### Post date: [April 19, 2018, 5:57pm UTC](https://discourse.julialang.org/t/using-survey-inverse-probability-weights-in-regression/10444/1 "2018-04-19T17:57:23Z")
Hi, I hope the answer is not too obvious as I am new to Julia.
I want to do some regressions with a weighted sample. At the moment I use the GLM package and I saw there are ProbabilityWeights, but I haven’t figured out how to use them together. Can anybody help?
---
### Author:  [@Nosferican](https://discourse.julialang.org/u/Nosferican)
#### Post date: [April 19, 2018, 8:18pm UTC](https://discourse.julialang.org/t/using-survey-inverse-probability-weights-in-regression/10444/2 "2018-04-19T20:18:14Z")
GLM offers limited support for weights at the moment. The basic usage goes like,
```julia
using DataFrames, GLM
srand(0)
df = DataFrame(y = rand(1:10, 10),
x = rand(10),
w = rand(1:10, 10))
glm(@formula(y ~ x), df, Normal(), IdentityLink()) # OLS
glm(@formula(y ~ x), df, Normal(), IdentityLink(),
wts = float.(df[:w])) # WLS
```
---
### Author:  [@jhartma](https://discourse.julialang.org/u/jhartma)
#### Post date: [April 19, 2018, 8:43pm UTC](https://discourse.julialang.org/t/using-survey-inverse-probability-weights-in-regression/10444/3 "2018-04-19T20:43:34Z")
Thanks, it works smoothely.
Compared to R’s survey package I noticed smaller standard errors. Do you have any idea why?
---
### Author:  [@nalimilan](https://discourse.julialang.org/u/nalimilan)
#### Post date: [April 19, 2018, 8:45pm UTC](https://discourse.julialang.org/t/using-survey-inverse-probability-weights-in-regression/10444/4 "2018-04-19T20:45:57Z")
GLM interprets weights as analytic (a.k.a. inverse variance) weights, like R’s `glm`. If you pass it sampling (a.k.a. inverse probability) weights, you’ll get incorrect standard errors.