Hello Analytics,
I read "Hiring a data scientist" ( https://blog.wikimedia.org/2017/02/02/hiring-data-scientist/) recently and it gave us a lot of useful insights into our (WMDEs) process of hiring a Data Analyst.
I stumbled upon the Bootstrap question and was reminded that I should extend my knowledge there; I read about the theory behind but I never used it in practice, so I wondered if one of you has some examples to share where you used bootstrapping in your work (Ideally in a iPython or RMarkdown notebook, which usefulness was also highlighted in the Job post :-) )
Jan
Hi Jan,
Glad to hear the post is useful to you! :D Here's a practical, self-contained example of bootstrapping: https://cran.r-project.org/web/packages/broom/vignettes/bootstrapping.html
And an excerpt from my own work: https://github.com/wikimedia-research/Discovery-Search-Test-PhraseRescoreBoo... because I needed to bootstrap the distribution of a K-L divergence metric ( https://en.wikipedia.org/wiki/Kullback%E2%80%93Leibler_divergence)
Also, if you are using the R package "forecast" for ARIMA models, one of the options in predict.Arima (I think) is to create confidence intervals for the forecast by bootstrapping residual errors :)
Hope that helps!
- Mikhail
On Saturday, February 4, 2017, Jan Dittrich jan.dittrich@wikimedia.de wrote:
Hello Analytics,
I read "Hiring a data scientist" (https://blog.wikimedia.org/ 2017/02/02/hiring-data-scientist/) recently and it gave us a lot of useful insights into our (WMDEs) process of hiring a Data Analyst.
I stumbled upon the Bootstrap question and was reminded that I should extend my knowledge there; I read about the theory behind but I never used it in practice, so I wondered if one of you has some examples to share where you used bootstrapping in your work (Ideally in a iPython or RMarkdown notebook, which usefulness was also highlighted in the Job post :-) )
Jan
-- Jan Dittrich UX Design/ User Research
Wikimedia Deutschland e.V. | Tempelhofer Ufer 23-24 | 10963 Berlin Phone: +49 (0)30 219 158 26-0 http://wikimedia.de
Imagine a world, in which every single human being can freely share in the sum of all knowledge. That‘s our commitment.
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e. V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/029/42207.