data science and statistical computing

0.0(0)
studied byStudied by 0 people
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
Card Sorting

1/71

encourage image

There's no tags or description

Looks like no tags are added yet.

Study Analytics
Name
Mastery
Learn
Test
Matching
Spaced

No study sessions yet.

72 Terms

1
New cards

unbiased estimate of the population standard deviation

knowt flashcard image
2
New cards

estimate of the standard deviation of the sample mean

knowt flashcard image
3
New cards

a t-distributed variable

knowt flashcard image
4
New cards

one sided p-value

knowt flashcard image
5
New cards

two sided p-value

knowt flashcard image
6
New cards
<p>monte-carlo testing</p><p>given data (x1,…,xn), an observed test statistic t = h(x1,….,xn), a data generating distribution F(x|theta) and a hypothesis as below, give the steps to perform a monte carlo hypothesis test.</p><p></p>

monte-carlo testing

given data (x1,…,xn), an observed test statistic t = h(x1,….,xn), a data generating distribution F(x|theta) and a hypothesis as below, give the steps to perform a monte carlo hypothesis test.

(known s.d. but not unknown)

<p>(known s.d. but not unknown)</p>
7
New cards

simply putting the process of hypothesis testing

knowt flashcard image
8
New cards
<p>bootstrap estimate and standard error</p><p></p>

bootstrap estimate and standard error

knowt flashcard image
9
New cards

cdf (or just distribution function)

knowt flashcard image
10
New cards

empirical cdf (or empirical distribution function)

knowt flashcard image
11
New cards

conditions for a valid cdf

knowt flashcard image
12
New cards

Glivenko-Cantelli theorem

knowt flashcard image
13
New cards

lemme 3.1 - sampling uniformly at random is equivalent to…

knowt flashcard image
14
New cards

ecdf approximates…

the true cdf

15
New cards

bootstrap resampling is equivalent to…

sampling n times from the ecdf (with replacement)

16
New cards

bootstrap is equivalent to first fitting an ecdf to the data and then…

sampling from it as though this was a fitted distribution

17
New cards

probability mass function when assigning probability 1/n at each value xi (for discrete random variable)

knowt flashcard image
18
New cards

E[Y]

knowt flashcard image
19
New cards

Var[Y]

knowt flashcard image
20
New cards

look at the mean of a sample of m draws from an ecdf constructed on n data points, and give the expectation and variance of the mean:

knowt flashcard image
21
New cards

bootstrap standard error for the mean

knowt flashcard image
22
New cards

special cases in which the standard bootstrap procedure needs more care

knowt flashcard image
23
New cards

what value of f = n/N indicates that the effect of the finite population size cannot be dismissed

0.1

24
New cards

theorem 3.2 - finite population variance of the mean

knowt flashcard image
25
New cards

important notes about the finite population variance of the mean

knowt flashcard image
26
New cards

theorem 3.3 - unbiased estimator of Var(Xbar)

knowt flashcard image
27
New cards

altering the finite case so that it works

knowt flashcard image
28
New cards
<p>population bootstrap</p>

population bootstrap

knowt flashcard image
29
New cards
<p>parametric bootstrap estimate and standard error</p>

parametric bootstrap estimate and standard error

knowt flashcard image
30
New cards

using bootstrap samples

knowt flashcard image
31
New cards

bias from stats I

knowt flashcard image
32
New cards

basic bootstrap bias estimate

knowt flashcard image
33
New cards

bias correction

knowt flashcard image
34
New cards

100(1-a)% confidence interval using the bootstrap estimate of standard error

where za/2 is the 100(a/2)% percentile of that standard normal distribution

<p>where za/2 is the 100(a/2)% percentile of that standard normal distribution</p>
35
New cards

100(1-a)% confidence interval

knowt flashcard image
36
New cards

probability of the sampled coordinates being inside the circle

knowt flashcard image
37
New cards

mu as an integral

knowt flashcard image
38
New cards

approximating mu (as a sum)

knowt flashcard image
39
New cards

weak law of large numbers

knowt flashcard image
40
New cards

expressing mu as an integral with Y=g(X)

knowt flashcard image
41
New cards

X ~ Unif(a,b) pdf

knowt flashcard image
42
New cards

when Var(Y) = σ² < inf, then give the expectation of μ^n

knowt flashcard image
43
New cards

Var(μ^n)

knowt flashcard image
44
New cards

mean squared error (MSE) and root mean squared error (RMSE)

knowt flashcard image
45
New cards

orders f functions

knowt flashcard image
46
New cards

Riemann integral

knowt flashcard image
47
New cards

using Chebyshev’s inequality to provide a probabilistic bound on the absolute error exceeding a desired tolerance

knowt flashcard image
48
New cards

CLT

knowt flashcard image
49
New cards

confidence interval where g(.) is some constant multiple of an indicator function

knowt flashcard image
50
New cards

p^n

knowt flashcard image
51
New cards

binomial confidence interval

knowt flashcard image
52
New cards

overall integral for mu

knowt flashcard image
53
New cards

probability inequality

knowt flashcard image
54
New cards

relative error at most δ

knowt flashcard image
55
New cards

Bayesian posterior and its expectations

knowt flashcard image
56
New cards

statement of probability computable as expectations

knowt flashcard image
57
New cards

inverse transform sampling

knowt flashcard image
58
New cards

generalised inverse cdf

knowt flashcard image
59
New cards

theorem 5.1 - generalised inverse cdf

knowt flashcard image
60
New cards

rejection sampling

knowt flashcard image
61
New cards

lemma 5.1 - when proposal an target are normalised pdfs…

knowt flashcard image
62
New cards

theorem 5.2 - generating X

knowt flashcard image
63
New cards

mu (importance sampling)

knowt flashcard image
64
New cards

importance sampling

knowt flashcard image
65
New cards

E[Xh(X)] = … and Ef[Xg(X)] = …

knowt flashcard image
66
New cards

theorem 5.3 - μhat

knowt flashcard image
67
New cards

theorem 5.4 - the variance of the importance sampling estimator

knowt flashcard image
68
New cards

optimal f tilda (x)

knowt flashcard image
69
New cards

self-normalised importance sampling

knowt flashcard image
70
New cards

approximate estimate and what is uses

knowt flashcard image
71
New cards

theoretical optimal proposal in the self-normalised weight case

knowt flashcard image
72
New cards

effective sample size

knowt flashcard image