Recap Flashcards

(14 cards)

1
Q

what is the difference between statistics and probability

A

Statistics reason backwards. Probability reasons forwards.

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

What is moments

A

Moments refer to how the probability mass is distributed.

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

how to find the k’th moment?

A

E[x^k]

E[(X - c)^k] = Integral {(x-c)^k f(x) dx} from -infinity to infinity

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

what is a raw moment?

A

A raw moment is a moment about the origin, c=0

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

what is a central moment?

A

moment about the mean of the distribution, where we therefore have c=E[X]

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

what are the 4 first moments

A

expecation
variance
skewness
kurtosis

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

how do we differ between statistics and estimates?

A

upper case for statistics, because they are random variables. Lower case for their corresponding estimates.

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

elaborate on sampling distributions

A

Consider a statistics. If we sample it, and do this many times, we will get some distribution which we refer to the sampling distribution.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

elaborate on MLE

A

We need the distribution of the variables we are working with. then we fidn the joint probablity densitriy. This is the likelihood function. then we log it because it is easier to work with. then we differnetiate partially for each parameter and solve for 0. this then will arrive at estimators for the population parameters.

It is key to udnerstand that MLE will arrive at estimators for the population parameters. So, if we are working with the normal distribution, we will get estimators for the mean and variance.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

how do we define a good estimatpr

A

We want the sampling distribution to be as concentrated around the true population parameter value as possible.

we have 3 properties:
1) Unbiasedness
2) efficiency
3) Consistency

If unbiased, the expected value opf the estimator yield the true population parameter.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

how do we define confidence intervals

A

P(value inside some region) = 1 - alpha

1-alpha is called the confidence level.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

what is type 1 error

A

rejecting the null hypothesis when it was really true

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

what is type 2 error

A

not rejecting the null when it is false

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly