[lab-reading-stats-concepts] Gonçalo Nobre

.ipynb_checkpoints/lab-reading-stats-concepts-checkpoint.ipynb

@@ -0,0 +1,6 @@
+{
+ "cells": [],
+ "metadata": {},
+ "nbformat": 4,
+ "nbformat_minor": 4
+}

lab-reading-stats-concepts.ipynb

@@ -0,0 +1,72 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Challenge 1: What is the difference between expected value and mean?\n",
+    "\n",
+    "Expected value is the average value of a random variable over a large number of experiments. We can calculate expected value for a discrete random variable — one in which the number of potential outcomes is countable — by taking a sum in which each term is a possible value of the random variable multiplied by the probability of that outcome.\n",
+    "\n",
+    "The mean is the average of a set of numbers. To find the mean of a data set, add up all of the numbers in the set, and then divide that total by the number of numbers in the set.\n",
+    "\n",
+    "A practical approach results in a frequency distribution and a mean value; a theoretical approach results in a probability distribution and an expected value.\n",
+    "If the sample space is infinitely large, the mean value should approach the expected value -> every time!"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Challenge 2: What is the \"problem\" in science with p-values?\n",
+    "\n",
+    "Researchers have been warned that a statistically non-significant result does not ‘prove’ the null hypothesis (the hypothesis that there is no difference between groups or no effect of a treatment on some measured outcome)1. Nor do statistically significant results ‘prove’ some other hypothesis.\n",
+    "\n",
+    "Like mentioned in the article: we should never conclude there is ‘no difference’ or ‘no association’ just because a P value is larger than a threshold such as 0.05 or, equivalently, because a confidence interval includes zero. Neither should we conclude that two studies conflict because one had a statistically significant result and the other did not.\n",
+    "\n",
+    "The problem in science is that bucketing results into ‘statistically significant’ and ‘statistically non-significant’ makes people think that the items assigned in that way are categorically different -> which it can or cannot be the case!"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Challenge 3: Applying testing to a specific case: A/B testing.\n",
+    "\n",
+    "Netflix example:\n",
+    "\n",
+    "Netflix knows that if you don’t capture a member’s attention within 90 seconds, that member will likely lose interest and move onto another activity.\n",
+    "Through various studies, they found that the members look at the artwork first and then decide whether to look at additional details, and they wanted to capitalize on this.\n",
+    "Broadly, Netflix’s A/B testing philosophy is about building incrementally, using data to drive decisions, and failing fast.\n",
+    "\n",
+    "They experimented using a movie -> 'The Short Game' measured the engagement with the title for each variant — click through rate, aggregate play duration, fraction of plays with short duration, fraction of content viewed (how far did you get through a movie or series), etc, changing the image from user to user,\n",
+    "\n",
+    "Then, they expanded to a  two way multi-cell explore-exploit test, where they measured the engagement of each user artwork for a set of titles -> \"explore\".\n",
+    "Finally they went to \"exploit\": test served the most engaging artwork (from explore test) for future users and see if we can improve aggregate streaming hours.\n",
+    "\n",
+    "In my opinion this is very useful since we can A/B test everything, at every time, getting real data to help us making the decisions!\n"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.7.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}