Probability Flashcards

You may prefer our related Brainscape-certified flashcards:
1
Q

Give the formula for

1) Conditional Probability
2) to check if a variable is independent to the other

A
P(A|B) = P(A^B) / P(B)  a different way to write it is:
P(A|B) = P(A,B) / P(B)

When E and F are independent, you can check that this gives: P(A|B] = P(A) which is the mathematical way of expressing that knowing F occurred gives us no
additional information about whether E occurred.

# | means given. So A givens B. So we investigate what happens with A if B is given.
# ^ means that both happen. Note: it is actually written down, not up.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Baye´s Theorem.

  1. Derive it from a decision tree and state it like it´s normally stated.
  2. What notation to use to say: not happens?
A
  1. P(A|B) = P(A)P(B|A) / P(B) # den decision tree Weg entlang wird multipliziert.
    • | (mittelstrich ist oben, sodass 90 Grad Winkel)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Continous Distributions

What do we do with a Probability Density Function (PDF) and what is it?

A

we represent a continuous distribution with a probability density function (pdf) such that the probability of seeing a value in a certain interval equals the integral of the density function over the
interval

f(x) represents the height of the curve at point x. Imagine the function as a smooth histogram, like a mountain landscape, where the area underneath is filled out.

for continous random variables probabilities are areas unter the curve.

The area under the entire curve is equal to one.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is a cumulative distribution function (cdf)?

A

While pdf describes the shape of the distribution, cdf describes the area in it (the accumulated probability), from a certain point (p.e. x) to the LEFT.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Name the 3 forms of pdf and cdf and their formulas

A

uniform:
Pdf= f(x)= 1/ b-a
CDF = AL= x-a/b-a #AL is Fläche, a linker Punkt, x mitte and b right

exponential:
Pdf= f(x)= 入e- 入x # 入 is highest point, * means hoch - 入x
CDF: 1-e
- 入x

standars:
other card

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Describe the Normal Distribution (Gaussian).

Name the formula!

A

by deault the Standard Deviation σ (sigma) is 1 and the mean μ (mu) is 0.
Mu shows where the bell is centered, sigma the width.
1 SD apart gets 68%, 2 SD gets 95, 3SD gets 99,7

Formula: 1 / σ √2π e(hoch)-(x-μ)2 / 2σ2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is a Bernouille trial?

A

In a Bernouille trial there is only success and failure. A coin flip is either heads or tails, so only two possible outcomes with the same probability. Every time the trial is conducted, the probabilities are the same.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What does a Confidence Interval do?

A
  • CI is a range of values where your assumption should fall in (p.e. if you think the mean will be 4.3 you can p.e. have a range of 2.6-5.9)
  • a way to measure how well your sample represents the population you are studying
  • The probability that the confidence interval includes the true mean value within a population is called the confidence level of the CI (normally by 95%)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is a p-value?

What is p-hacking?

A

Tells us if the observations can support the 0-Hypothesis. Is p-value under 0,05 you can reject the 0-Hypothesis, and thus say that the alternative Hypothesis is true. If that happens - throwing 0 away for alternative, it is called statistical significance

Misinterpretation and intentionally pushing the value under 0,05 to increase the chance of getting published is called p-hacking

How well did you know this?
1
Not at all
2
3
4
5
Perfectly