layout: true .absolute.top-0.right-1.tr.w-10[ ![](https://raw.githubusercontent.com/jmbuhr/dataIntro20/master/images/hex.png)<!-- --> ] --- name: title class: left bottom hide-count background-color: #FBFCFF;
Introduction to Data Analysis with R
Lecture 6: Continuous Distributions, Statistics and PCA
Jannik Buhr
Heidelberg University, WS20/21
2020-12-07
.absolute.bottom-0.right-1.mid-gray[ With Artwork by @allison_horst ] --- class: center, middle ## Reproducible Environments with `renv` <a href="https://rstudio.github.io/renv/"> <img src="img/renv.svg" width="40%"> </a> --- class: middle > »All models are wrong, but some are useful« > — George Box --- ## Types of Models 1. **Descriptive Models** 2. **Inferential Models** 3. **Predictive Models** <img src="slides6_files/figure-html/unnamed-chunk-2-1.png" width="864" /> --- ## Boxplots <img src="lecture6_files/figure-html5/box-plot-outlier-1.png" width="832" /> --- ## Mean, Variance and SD `$$var(X) = \frac{\sum_{i=0}^{n}{(x_i-\bar x)^2}}{(n-1)}$$` `$$\sigma_X=\sqrt{var(X)}$$` <img src="lecture6_files/figure-html5/valence-mean-1.png" width="555" /> --- ## Standard Error of the Mean `$$SEM=\sigma / \sqrt{n}$$` --- <img src="lecture6_files/figure-html5/horrible-plot-1.png" width="832" /> --- So the next time you see a barplot ask the question: ![[@ArtworkAllisonHorst]](img/summary_statistics.png){.external} --- ## The t-distribution <img src="lecture6_files/figure-html5/tdist-1.png" width="555" /> --- background-image: url(img/whaleshark.png) background-size: contain --- background-image: url(img/krill.png) background-size: contain