Posts
Note: If a concept is new to you, follow the link for my explanation. If the early stuff feels too technical, feel free to skip to the cuddly critter memes lower down.
 |
| Get your distribution basics in Part 1 if you’re new to this space. Image: SOURCE. |
Ready for the list of favorites? Let’s dive right in!
Take a “moment” to explore some fundamentals.
This article takes you on a tour of the most popular parameters in statistics! says Cassie Kozyrkov, Head of Decision Intelligence, Google, published in Towards Data Science.
If you’re not sure what a statistical parameter is or you’re foggy on how probability distributions work, I recommend scooting over to my beginner-friendly intro here in Part 1 before continuing here.
MeanExpected value
An expected value, written as E(X) or E(X=x), is the theoretical probability-weighted mean of the random variable X.
You find it by weighting (multiplying) each potential value x that X can take by its corresponding probability P(X = x) and then combining them (with an integral for continuous variables like height or a sum for discrete variables like height-rounded-to-the-nearest-inch): E(X) = ∑ x P(X=x)
If we’re dealing with a fair six-sided die, X can take each value in {1, 2, 3, 4, 5, 6} with equal probability 1/6, so:
E(X) = (1)(1/6) + (2)(1/6) +(3)(1/6) +(4)(1/6) +(5)(1/6) +(6)(1/6) = 3.5
In other words, 3.5 is the probability-weighted average for X and nobody cares that 3.5 isn’t even an allowable outcome of the die roll.
VarianceFor reasons that I’ll explain in a moment, replacing X with (X — E(X))² in the E(X) formula above gives you something very useful.
Read more...
Source: Medium
.jpg){kind=link}