Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

CS 3341 > Re-Study > Flashcards

Re-Study Flashcards

1
Q

In a long run, probability can be viewed as what?

A

The proportion of times an event happens, or its relative frequency.

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

What is a sample space?

A

A collection of all elementary results, or outcomes of an experiment.

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

What is an event?

A

Any set of outcomes, and a subset of the sample space.

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

A sample space of N possible outcomes yields how many possible events?

A

2n possible events.

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

What is the notation for the sample space?

A

The Capital Omega

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

What is the notation for the empty event?

A

Ø

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

What is the notation for the probability of the event E?

A

P{E}

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

A union of events A, B, C, is an event consisting of what? What word does this correspond to?

A

all the outcomes in all these events. It corresponds to the word or.

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

A complement of an event A is what? What word does it correspond to?

A

an event that occurs every time when A does not occur. It corresponds to the word not.

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

An intersection of events A, B, C… is what and corresponds to what word?

A

an event consisting of outcomes that are common in all these events. It occurs if each A, B, C, … occurs, and therefore corresponds to the word and.

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

A difference of events A and B consists of what, and corresponds to what phrase?

A

all outcomes included in A but excluded from B, and corresponds to the words “but not.” A but not B.

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

Events A and B are disjoint if

A

their interesection is empty

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

If any two events are disjoint in a set of events, they are?

A

Mutually exclusive

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

Another term for mutually exclusive

A

Pairwise disjoint

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Events A B and C are exhaustive if

A

their union equals the whole sample space.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Occurrence of a mutually exclusive event does what?

A

Eliminates the chance of any other mutually exclusive event occuring.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

A single event A and it’s compliment is a classical example of what?

A

A collection of disjoint, and exhaustive events.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

If a collection of events is exhaustive then

A

One event must occur.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

The compliment of a union of two events is

A

the intersection of the compliments of both events.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Notation for the difference of A and B

A

A/B

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What is the sigma-algebra?

A

a collection of events whose probabilities we can consider in our problem.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What makes a collection of events a sigma-algebra on a sample space?

A

It includes the sample space.

It includes every event, and its compliment.

Every coutable collection of events in the sigma-algebra is contained along with their unions.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

What is the minimal collection of a sigma-algebra?

A

The sample space, and the empty event.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What is the minimal collection of evens for a sigma-algebra known as?

A

The degenerate Sigma-algebra.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

What is the power set of the sigma algebra, and what is its size?

A

The collection of all events and their unions. Its size is 2Omega

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

What is the sigma additivity problem?

A

for any finite or countable collection of mutually exclusive events, P{E1 U E2…..} = P(E1) + P(E2)…

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

What is the formal definition of probability?

A

Probability is a function of events with the domain sigma-algebra and the range [0,1] that satisfies the sigma-additive property, and the sample space has unit probability or P(sampel space) = 1.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

What is the probability of an empty event?

A

0

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

The probability of an event is equal to what?

A

The sum of all of the mutually exclusive outcomes contained in that event.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

Only what kind of events satisfy the Sigma-additivity property?

A

Mutually exclusive events.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

How do you calculate the probability of events that are not mutually exclusive?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
32
Q

What is the compliment rule?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
33
Q

How do you calculate the probability of independent events?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
34
Q

When are events independent?

A

When the occurence of one event does not affect the probabilities of other events occuring.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
35
Q

What is the notation for the sigma algebra?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
36
Q

What is a random variable?

A

A variable that depends on chance.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
37
Q

What is a stochastic process?

A

A experiment model in which the random variables depend on time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
38
Q

What is fundamental to correctly determining the likelihood of an experiment’s outcomes?

A

Precisely defining the experiment is fundament to determining

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
39
Q

When can we say for certain the value of a random variable?

A

We can’t, we can only talk about the distribution or all possible values of a random variable with the likelihood of occurence.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
40
Q

What are the three interpretations of probability?

A

Classical, subjective or bayesian, and frequentist

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
41
Q

What is the classical interpretation of probability?

A

We have an intuitive idea of probability and in some situations already know how to compute it. Such as rolling a 6 sided dice with equally likely outcomes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
42
Q

What is the frequentist interpretation of probability?

A

We have an intuitive idea of probability in some situations that we do not compute on our own, but is based on past observations.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
43
Q

What is the subjective or Bayesian interpretation of probability?

A

Probability is a degree of belief. We have an intuitive idea of probability that may not fit the classical or frequentist interpretations.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
44
Q

What is an example of Bayesian interpretation of probability?

A

One in which the experiment can not be made, it is destructive. What is the probability of that bridge collapsing?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
45
Q

When do we say an event occurred?

A

When the outcome of an experiment is a member of that event.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
46
Q

An experiment will have how many outcomes?

A

Exactly one.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
47
Q

What does it mean that ¬A is relative to the sample space?

A

It means that ¬A includes everything in the sample space not in A.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
48
Q

What is the probability of each outcome when the sample space consists of n equally likely outcomes?

A

1/n

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
49
Q

How is the probability of an event calculated?

A

(# of out comes in event)/(# of outcome in sample space) * for equally likely outcomes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
50
Q

In reality most situations do not have what?

A

Do not have equally likely outcomes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
51
Q

Equally likely outcomes are usually associated with the phrases

A

“fair game” or randomly selected.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
52
Q

Outcomes forming an event are often called what?

A

Favorable outcomes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
53
Q

What does sampling with replacement mean?

A

means that every sampled item is replaced into the initial set, so that any of the objects can be selected with probability 1/n at any time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
54
Q

What provides special techniques for the computation of favorable outcomes and total outcomes?

A

Combinatorics.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
55
Q

When sampling with replacement, the same object may what?

A

Be sampled more than once.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
56
Q

What does sampling without replacement mean?

A

every sampled item is removed from further sampling, so the set of possibilities reduces by 1 after each selection.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
57
Q

When are objects distinguishable?

A

if sampling of exactly the same objects in a different order yields a different outcome, that is, a different element of the sample.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
58
Q

When are objects indistinguishable?

A

if the order is not important, it only matters which objects are sampled and which ones are not. Indistinguishable objects arranged in a different order do not generate a new outcome.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
59
Q

What is an example of is an example of distinguishable objects without replacement?

A

A password.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
60
Q

How are permutations with replacement calculated?

A

Where n is the possible selections, and k is how many selections.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
61
Q

How are permutations calculated without replacement?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
62
Q

What are permutations?

A

Possible selections of k distinguishable objects from a set of n are called

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
63
Q

How do you calculate combinations without replacement?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
64
Q

The numbe of permutations of k, n is equal to what?

A

the number of possible allocations of k distinguishable objects among n available slots.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
65
Q

What are combinations?

A

Possible selections of k indistinguishable objects from a set of n

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
66
Q

What is an example of a combination?

A

An antivirus software reports that 3 folders out of 10 are infected, how many possibilities are there? Order in this case does not matter, A, B, C is the same outcome as B, A, C.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
67
Q

What is conditional probability?

A

event A given event B is the probability that A occurs when B is known to occur.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
68
Q

How is conditional probability denoted?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
69
Q

How is Conditional probability of A given B calculated?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
70
Q

How can Conditional probability of A given B be simplified to give us the probability of the general intersection?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
71
Q

How can independence be mathematically defined?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
72
Q

How do we know if events are independent?

A

If

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
73
Q

is the probability of A given B equal to the probability of B given A?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
74
Q

What can be used to find

A

Bayes Rule

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
75
Q

What is Bayes Rule?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
76
Q

What is independence?

A

Events A and B are independent if occurrence of B does not affect the probability of A

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
77
Q

In the case of two conditional events A and B how is the probability of A calculated using the law of total probability?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
78
Q

How is Bayes rule for two events calculated using the law of total probability for A?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
79
Q

What is often used to calculate the denominator in Bayes Rule?

A

The law of total probability.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
80
Q

What does the law of total probability do?

A

It relates the unconditional probability of an event A with its conditional probabilities

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
81
Q

When is law of total probability used?

A

when it is easier to compute conditional probabilities of A given additional information.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
82
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
83
Q

A random variable X is a function of what?

A

It is the function of an outcome σ of an experiment, X = f(σ), in other words it is a variable that depends on chance. We can not know what X is until an experiment has an outcome.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
84
Q

What is the domain of a random variable.

A

The sample space is its domain.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
85
Q

What is the range of a random variable?

A

It an be the set of all real number or any subset of the real numbers, only dependent on what values a random variable can take.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
86
Q

When working with a random variable X, what do we chart?

A

We chart all of the possible values x, and their corresponding probabilities.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
87
Q

What is known as the distribution of X?

A

The collection of all probabilities related to X.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
88
Q

What is the set of all possible values of X called?

A

The support of the distribution.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
89
Q

What is the cumulative distribution function?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
90
Q

What is the probability mass function of a value x?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
91
Q

What are discrete random variables?

A

variables whose range is finite or countable.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
92
Q

What A is an inteval from a to b, how can its probability be computed directly from the cumulative distributive property?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
93
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
94
Q

What is the set

A

exhaustive and mutually exclusive events for different pairs (x, y).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
95
Q

What is the addition rule for when using two random variables?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
96
Q

When are two random variables independent?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
97
Q

What are continuous random variables?

A

variables whose range assume a whole interval of values. This could be a bounded interval (a, b), or an unbounded interval .

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
98
Q

Expected value is denoted with what?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
99
Q

What is the general formula for the expectation?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
100
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
101
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
102
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
103
Q

What is an example of a continous random variable?

A

A long jump is formally a continuous random variable because an athlete can jump any distance within some range.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
104
Q

How is the variance of a random variable calculated?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
105
Q

When does the variance equal zero?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
106
Q

What is the expectation of a random variable?

A

its mean, the average value

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
107
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
108
Q

How is the correlation coefficient calculated?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
109
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
110
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
111
Q

What is Chebyshev’s inequality?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
112
Q

Suppose the number of error in a new software has Exp(X) = 20, and the standard deviation of 2, the probability of the software having more than 30 errors is

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
113
Q

If X and Y are integers what can their expectations be?

A

Any real number.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
114
Q

What does expectation show?

A

where the average value of a random variable is located, or where the variable is expected to be, plus or minus some error.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
115
Q

How is the variability of a random variable’s value measured?

A

Measured by its distance from the Expectation.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
116
Q

How is standard deviaton denoted?

A

σ

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
117
Q

How is standard deviation calculated?

A

It is the square root, +/-, of the Variance.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
118
Q

What is covariance?

A

summarizes interrelation of two random variables.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
119
Q

What does it mean if Cov(X, Y) = 0?

A

There is no correlation between the two variables.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
120
Q

What does the correlation coefficient do?

A

tells how strongly two variables are correlated, values near 1 indicate strong positive correlation, values near -1 show strong negative correlation, and values near 0 show weak correlation or no correlation.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
121
Q

For independent X and Y, Cov(X, Y) equals what?

A

It equals zero.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
122
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
123
Q

The probability of at least 2 is the compliment of what?

A

1 or less, at most 1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
124
Q

What is the probability of Event 1 and Event 2 when Event 2 is a member of Event 1

A

The probability is the probability of Event 2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
125
Q

What is a Bernoullie variable?

A

A random variable that can only take on two possible values, 0 and 1.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
126
Q

What is a Bernoulli trial?

A

An experiment with a binary outcome.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
127
Q

What are some examples of Bernoulli trials?

A
  • Pass or fail tests
  • Heads, or tails
  • Boys or girls
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
128
Q

What are the two generic names used for the outcomes of Bernoulli trials?

A

Successes and Failures; however successes do not have to be good and failures do not have to be bad.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
129
Q

In a Bernoulli distribution, if P(1) = p, what is P(0)?

A

1-p

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
130
Q

The expectation of a Bernoulli trial is always what?

A

The proability of a success.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
131
Q

The variability of a Bernoulli variable is always what?

A

the product of the probabilities of succes and failure.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
132
Q

The number of Bernoulli trials needed to get the first success has what kind of distribution?

A

It has geometric distribution

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
133
Q

What is an example of an experiment with geometric distribution?

A

A search engine goes through a list of sites looking for a given key phrase, and terminates as soon as the key phrase is found. The number of sites visited is geometric.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
134
Q

Geometric Random Variables can take what?

A

Any integer value from one to infinity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
135
Q

What is the probability mass function for a geometric distribution?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
136
Q

How is the probability of a Binomial distribution described?

A

How many success in n trials.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
137
Q

What kind of Variable has Binomial distribution?

A

A variable described as the number of successes in a sequence of independent Bernoulli trials.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
138
Q

What is the PMF of a Binomial distribution?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
139
Q

What do works like least and most usually mean?

A

They usually mean that the CDF should be sought for.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
140
Q

What table has the CDF of Binomial Distrubutions?

A

Table A2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
141
Q

What is the expectation of a binomial distribution?

A

np, where n is the number of trials.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
142
Q

What is the variance of a Binomial distribution?

A

npq, where q is 1-p

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
143
Q

What is the geometric distribution?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
144
Q

What has negative binomial distribution

A

the number of trials needed to obtain k successes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
145
Q

What are Poissonian events?

A

events that are extremely unlikely to occur simultaneously or within a very short period of time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
146
Q

Binomial varibles count what?

A

the number of successes in a fixed number of trials

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
147
Q

Negative Binomial variables count what?

A

the number of trials needed to see a fixed number of successes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
148
Q

What is the Poisson distibution?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
149
Q

What has Poisson distibution?

A

The numer of rare events occuring within a fixed period of time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
150
Q

What are examples of Poissonian events?

A

traffic accidents, arrivals of jobs, telephone calls, virus attacks, floods, and earthquakes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
151
Q

What table has the values of CDFS of Poissonian distributions?

A

Table A3

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
152
Q

If the period of time changes in a problem using Poisson distribution what needs to be adjusted?

A

only the frequency to what the average would be over the new time period.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
153
Q

what is the Expectation of a negative binomial distribution?

A

k/p where k is how many successes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
154
Q

What is the variance of a negative binomial distribution?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
155
Q

How do you calculate the PMF of a negative binomial distribution? part 1

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
156
Q

How do you calculate the PMF of a negative binomial distribution? part 2

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
157
Q

How do you calculate the PMF of a negative binomial distribution? part 3

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
158
Q

What is Poisson Approximation of a Binomial distribution?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
159
Q

For all continuous variables, P(x) = ?

A

zero.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
160
Q

In both continuous and discrete cases, the CDF is what?

A

a non-decreasing function that ranges from 0 to 1.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
161
Q

What is different about the CDF with continuous variables from discrete variables?

A

The CDF is a continuous function, and there are no jumps in the CDF.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
162
Q

With continuous variables, probabilities are what?

A

Areas under a density curve.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
163
Q

What is the probability density function?

A

A derivative of the CDF, f(x) = F’(X)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
164
Q

What is the total area under a pdf equal to?

A

The total area under a pdf is equal to 1.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
165
Q

What are the four families of continuous distributions discussed in this chapter?

A

Uniform, Exponential, Gamma, and Normal.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
166
Q

When is Uniform distribution used?

A

In any situation when a value is picked at random from a given interval.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
167
Q

Uniform distribution has constant what?

A

Density.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
168
Q

What is the density function for uniform distribution?

A

f(x) = 1/(b-a)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
169
Q

What must be true for use of a Uniform distribution?

A

|b-a| must be a finite countable number

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
170
Q

What does [a, b] represent in uniform distribution?

A

the domain of the uniform density function.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
171
Q

What is an example of a situation with uniform density?

A

If a flight is scheduled to arrive at 5pm actually arrives at a Uniformly distributed time between 4:50 and 5:10, then it is equally likely to arrive before five and after five.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
172
Q

What is the uniform property?

A

the probability is only determined by the length of the interval, not by its location.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
173
Q

How is the variance of a continuous random variable calculated?

A

\int x^2f(x)dx - E(x)^2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
174
Q

How is the expectation of a continuous variable calculated?

A

\int xf(x)dx

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
175
Q

What is the Uniform Distribution?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
176
Q

What is exponential distribution often used for?

A

To model time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
177
Q

What are some examples of exponential distributions?

A

waiting time, interarrival time, hardware lifetime

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
178
Q

When is the time between events exponential?

A

when the number of events is Poisson

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
179
Q

if X is time, measured in minutes, what is lambda?

A

The frequency, number in a time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
180
Q

If arrival occurs every half minute what is the expectation?

A

E(X), so it is .5, we expect to get one every .5 minutes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
181
Q

If arrivals occur every half a minute what is lambda?

A

lambda is 1/.5, so 2.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
182
Q

What does it mean that Exponential variables are memoryless?

A

It means that having waited for t minutes gets “forgotten,” and does not affect the future waiting time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
183
Q

The time T until the next event is greater than t can be rephrased as what?

A

Zero events occur by the time t.

184
Q

What is the exponential distribution?

A
185
Q

When does total time have gamma distribution?

A

When a certain procedure consists of alpha independent steps that each takes exponential amounts of time.

186
Q

In a process of rare events, what has gamma distribution?

A

In a process of rare events, with exponential times between any two consecutive events, the time of the ath event has Gamma Distribution.

187
Q

When does a gamma distribution become exponential?

A

When a =1.

188
Q

What is the Gamma Distribution?

A
189
Q

How can we significantly simplify gamma probabilities?

A

By thinking of a Gamma variable as the time between some rare events.

190
Q

What is the gamma poisson formula?

A

For a Gamma(a , lambda) variable T can be modeled with Poisson(lambda/t) where P{T>t} P{X < alpha} and P{T<= t} = P{X >= alpha}.

191
Q

What is normal distribution often found to be good for modeling?

A

sums, averages, errors, and physical variables like weight, height, and temperature.

192
Q

What is the CDF of the normal distribution?

A
193
Q

How do you normalize a Random variable X to Z?

A
194
Q

What is the full formula for Variance of X?

A

Var(X) = E[(X-E(X)^2)] = E(X^2) - E(X)^2

195
Q

Var(kX + C) = what?

A

K^2Var(X)

196
Q

What is the formula for covariance?

A

Cov(X,Y) = E[(X-E(X))(Y-E(Y))] = E(XY) - E(X)E(Y)

197
Q

What is the Variance of the sum of two variables?

A

Var(X+Y) = Var(X) + Var(Y) + 2Cov(X,Y)

198
Q

Variance is always expressed in what?

A

Squared units.

199
Q

What is the correlation between two variables?

A

Cov(X,Y)/(std(X)*std(Y))

200
Q

The correlation between two variables will always lie between what?

A

-1 and 1

201
Q

Let X be the percentage change in value of investment A in the course of one year, and let Y be the percentage change in value of investment B. You have 1 dollar. You invest a in A and 1-a in B. What is the return on investment?

A

Return: aX+(1-a)Y

202
Q

Let X be the percentage change in value of investment A in the course of one year, and let Y be the percentage change in value of investment B. You have 1 dollar. You invest a in A and 1-a in B. What is the expected return?

A

Expected Return: aE(X) + (1-a)E(Y)

203
Q

Let X be the percentage change in value of investment A in the course of one year, and let Y be the percentage change in value of investment B. You have 1 dollar. You invest a in A and 1-a in B. What is the variance in your return on investment?

A

a^2Var(X) + (1-a)^2Var(Y) + 2a(1-a)Cov(X,Y)

204
Q

What can a continuous random variable model?

A

Any analog measurements can be the domain of continuous random variables.

205
Q

Why can’t we sum all of the probability measures of a continuous random variable?

A

We can not enumerate all of the possible values a continuous random variable can take, so we have to integrate.

206
Q

What breaks down for continuous random variables, and what does not?

A

The PMF breaks down, but the CDF does not.

207
Q

Let X be a continuous random variable. What is P{X = x}?

A

0

208
Q

For continuous random variables, P{ X <= x} is the same as what?

A

It is the same as strictly less, P{X < x}. This goes for all states a < X and X > a, and a < X < B

209
Q

What is the powder description of PDF?

A

PDF is not PMF, it describes only density. If we were to break chalk up into a power and distribute it along a number line of where values that a continuous random variable is more likely to take, the denser areas would be where the likeliest values are located. A PDF curve represents this graphically. The CDF is a measure of the density within that region, as though we were to take the power in that region and measure it, but if we were to only measure the weight of one infinitesimal value, it would be zero.

210
Q

For a continuous random variable, P{a < X < b} is equal to what?

A

where f(x) is the pdf.

211
Q

P{X < b} = ?

A
212
Q

What are the two conditions for a function to be a legitimate PDF?

A
213
Q

Given f_x(X) = { cx^2 for 0 < x < 1 and zero other wise} Find c.

A
214
Q

Given f_x(X) = { 3x^2 for 0 < x < 1 and zero other wise} Find the cdf.

A

x^3

215
Q

What is the cdf of f(x)?

A
216
Q

When we are modeling continuous random variables, how is the probability density function denoted?

A
217
Q

What is the probability outside of the region enclosed by the pdf?

A

0

218
Q

P(X > x) = ?

A

P(X> x) = 1 - P(X < x) = 1-F(x)

219
Q

Without knowing a random variable’s distribution, what can we do if we know its parameters?

A

We can use Chebyshev’s inequality.

220
Q

What parameters must we know to use Chebyshev’s inequality?

A

The variance(or standard deviation) and the expectation

221
Q

How do we define the joint cumulative distribution function of two continuous random variables?

A
222
Q

How is the joint probability density function of a continuous random variable defined?

A
223
Q

How do we obtain the marginal pdf of a continuous random variable from a joint pdf?

A

We integrate the joint pdf over the random variables we want to eliminate.

224
Q
A
225
Q

Given a joint pdf, how can we find P{{a < X < b} and {a < Y < b}}?

A
226
Q
A
227
Q
A
228
Q

When we are speaking of uniform distribution, what does a larger value of b-a mean?

A

Larger variance

229
Q

Derive the Expectation and variance of the uniform distribution.

A

Do it

230
Q

What is the pdf of the exponential distribution?

A
231
Q

Derive the CDF, Expectation, and Variance of the exponential distribution.

A

do it

232
Q

How is the exponential distribution related to the Poisson Process?

A

The expectation of the exponential distribution is the reciprocal of the expectation of the poisson expectation. The time between Poisson processes is exponentially distributed.

233
Q

What does lambda represent?

A

lambda is the average number of events in some time interval.

234
Q

What does the memoryless property mean?

A

The past doesn’t matter, only the present. but more formally: where t is our time of arrival.

235
Q

Derive P{T>t}

A
236
Q

What is the only continuous distribution that has the memory less property?

A

The exponential distribution

237
Q

If we redefine lambda as the average number of events per Unit time and assume a period of interest as [0, t], what is the average number of events in that period?

A

t*lambda

238
Q

Gamma distribution is used to describe what?

A

times between events of Poisson process ( a process in which events occur continuously and independently at a constant rate).

239
Q

What is the gamma pdf?

A
240
Q

What is the gamma function?

A
241
Q

What is the expectation and variance of the Gamma distribution?

A
242
Q

What can we learn from the graph of a PDF?

A

The variance translates to the wideness of the PDF. The more “spread” out the PDF, the more values a random variable can take.

243
Q

What else is the Normal distribution known as?

A

the Gaussian distribution

244
Q

What is the mathematical representation of the probability that a random variable is within the neighborhood of 2 standard deviations?

A
245
Q

Besides 1-F(t), how can P{T>t} be calculated for Exponential modeling?

A
246
Q

What does the Poisson formula really mean at its heart?

A

It means that if less than this many events happens in this interval of time, then T > t for the gamma random variable.

247
Q

When are two continuous random variables independent?

A

If and only if their joint PDF and/or CDF can be factored into two individual pdf’s one of x and one of y, and the conditions on x and y remain the same.

248
Q

What does it mean mathematically if X and y are independent continuous random variables?

A

E(XY) - E(X)E(Y) = 0, or E(XY) =E(X)E(Y)

249
Q

How do we find the expectation of a continuous random variable?

A
250
Q

How do we find the variance of a continuous random variable?

A
251
Q

How do we find the covariance of a continuous random variable?

A
252
Q
A
253
Q
A
254
Q
A
255
Q

What is a stochastic process?

A

A random variable that also depends on time.

256
Q

A stochastic process is a function of what?

A

Two arguments X(t,w) time and the result. w exists in the sample space and is an outcome of an experiment

257
Q

What are values of X(t, w) called?

A

These are states

258
Q

What do we get if we fix a time of a Stochastic Process?

A

We get the function of an outcome: Xt(w)

259
Q

What is a realization or trajectory of a sample process?

A

It is a fixed outcome where we obtain a function of time Xw(t)

260
Q

When is a stochastic process discrete-state?

A

When Xt(w) is discrete for each time t

261
Q

When is stochastic process continuous state?

A

When Xt(w) is continuous for each time t.

262
Q

When is a stochastic process a discrete-time process?

A

When the set of times T, are discrete, that is consists of separate, isolated points.

263
Q

The CPU usage process in percent is what kind of stochastic process?

A

A continuous time, continuous state process.

264
Q

The temperature reported every hour and rounded to the nearest integer is what kind of process?

A

A discrete time, discrete state process.

265
Q

When is a stochastic process a continuous time process?

A

When T is a connected and possibly unbounded interval.

266
Q

If a stochastic process is markov, the conditional distribution of X(t) is the same under what two conditions?

A
  1. given observations of the process X at several moments in the past. 2. given only the latest observation of X.
267
Q

If a process is markov P{future |past, present} = ?

A

P{future|present}

268
Q

What is a Markov chain?

A

A discrete-time, discrete state Markov stochastic process.

269
Q

When is a Markov chain homogeneous?

A

When all its transition probabilities are independent of t. Being homogeneous means that transition from i to j has the same probability at any time.

270
Q

What does the Markov property mean?

A

It means that on the value of X(t) matters for predicting X(t+1)

271
Q
A

It is the probability of moving from state i to state j by means of h transitions.

272
Q

What is an h-step transition probability?

A

It is the probability of moving from state i to state j by means of h transitions.

273
Q

The distribution of a Markov chain is completely determined by what?

A

The initial state X(0) and one-step transition probabilities.

274
Q

What is our long term forecast?

A

The limit of our h-step transition probability.

275
Q

In some town, each day is either sunny or rainy. A sunny day is followed by another sunny day with probability 0.7, whereas a rainy day is followed by a sunny day with probability .4, what is the transition probability?

A

p11 = .7, p12= .3, p21 = .4, p22 = .6

276
Q

All one-step transition probabilities can be conveniently written as what?

A

An nXn matrix

277
Q

What do the rows represent in a transition probability matrix?

A

The from state

278
Q

What do the columns represent in a transition probability matrix?

A

The to state.

279
Q

What in a one-step transition probability matrix sums to 1?

A

Each row, but this is generally not true for the column totals.

280
Q

How do we fin the h-step transition probability matrix?

A

We raise the matrix P to the h power

281
Q

A computer is shared by 2 users who send tasks to a computer remotely and work independently. At any minute, any connected user may disconnect with probability 0.5, and any disconnected user may connect with a new task with probability 0.2. Let X(t) be the number of concurrent users at a time t. What are the states of the Markov chain?

A

0, 1, 2

282
Q

A computer is shared by 2 users who send tasks to a computer remotely and work independently. At any minute, any connected user may disconnect with probability 0.5, and any disconnected user may connect with a new task with probability 0.2. Let X(t) be the number of concurrent users at a time t. What is the transition probability matrix ?

A
283
Q

If the following is the 2step transition probability matrix, what is the probability the system will go from 2 users to 0 users after 2 units of time?

A

0.4225

284
Q

If we have n independent variables X with the same expectation and standard deviation, what can we use to predict their sum?

A

The central Limit Theorem

285
Q

If we have n independent variables X with the same expectation and standard deviation, what is the standardized sum X+X+X+X?

A
286
Q
A
287
Q

What are the conditions for using the central Limit theorem?

A

It applies to any random variables of virtually any thinkable distribution so long as they have the same finite expectation and variance, and as long as n is large > 30.

288
Q

When a large number of random variables that are independent and have the same expectation and standard deviation, the sum of those random variables create what?

A

A new random variable with Normal distribution.

289
Q

What kind of curve is the normal density curve?

A

A bell shaped curve, symmetric, and centered around the expectation.

290
Q

What does normalizing a random variable have the effect of?

A

Taking the bell curve and placing its expectation around zero and giving it a standard deviation of 1.

291
Q

The spread of the normal density curve is controlled by what?

A

The standard deviation.

292
Q

What does Z represent?

A

A standard normal random variable.

293
Q

What is the standard normal distribution?

A

It is a normal distribution with the expectation = 0, and std = 1.

294
Q

A disk has free space of 330 megabytes. Is it likely to be sufficient for 300 independent images, if each image has expected size of 1 megabyte with a standard deviation of .5 megabytes?

A

Each image is a random variable with its own expectation and standard deviation, so we use the central limit theorem.

295
Q

What is the formula for Normal approximation to Binomial Distribution?

A
296
Q

What is the continuity correction, when is it used?

A

It is used when we approximate a discrete distribution with a continuous distribution. P(x) = P{X = x} = P{x-0.5 < X < x+0.5}

297
Q

How do we obtain a standard normal variable from a nonstandard normal variables?

A
298
Q

How does changing the expectation to a normal distribution affect its density curve?

A

Changing the expectation shifts the curve to the right or to the left.

299
Q

How can we unstandardize Z?

A
300
Q

How is table A4 read?

A

The rows represent the first two digits of Z and the column represents the third digit of Z

301
Q

What is the probability density function of a Cauchy random variable?

A
302
Q

A cauchy random variable does not have what?

A

An expectation.

303
Q

What is the Cauchy distribution CDF?

A
304
Q

When a distribution is know known, what can Chebyshev’s inequality give us?

A

A bound

305
Q

modelling experiments as random variables describes what?

A

A single outcome of the experiment.

306
Q

A random process is a function that assigns what to what?

A

A time function to every outcome of a random experiment.

307
Q

We have a stochastic process if the outcome of an experiment results in what?

A

A function of time.

308
Q

What is Chebyshev’s inequality?

A
309
Q

If the outcomes of a stochastic process are continuous, but the Times are discrete, we have what?

A

A continuous random sequence.

310
Q

If the outcomes of a stochastic process are discrete and the time is continuous, what kind of process do we have?

A

A discrete random process.

311
Q

When do we have a continuous random process?

A

When we have a stochastic process in which the outcomes are continuous and so is the time.

312
Q

What does the autocorrelation of a stochastic process tell us?

A

It tells us how correlated an outcome at t2 is dependent on an outcome at t1.

313
Q

How can a distribution of states be represeted?

A

As a 1x n matrix or a row vector.

314
Q

How do we find the distribution of states after h transitions using matrix algebra?

A

Ph = P0P^h, P0 represents the distribution at X = 0, it can be a state distribution (0, 0, 1) or a distribution of probabilities (1/3, 1/3, 1/3). After matrix multiplication, we end up with a vector that represents our final distribution.

315
Q

How do we find the distribution of states after h transitions using matrix algebra?

A

Ph = P0P^h, P0 represents the distribution at X = 0, it can be a state distribution (0, 0, 1) or a distribution of probabilities (1/3, 1/3, 1/3). After matrix multiplication, we end up with a vector that represents our final distribution.

316
Q

What is a steady-state distribution?

A

A collection of limiting probabilities.

317
Q

A fast system will go through a very large number of transitions very quickly, its distribution of state is what?

A

A steady state distribution.

318
Q

What is the steady state formula?

A

piP = pi

319
Q

All pie in the steady state distribution must add to what?

A

They must add to 1.

320
Q

The steady state distribution is the solution to what?

A

The system piP = pi, and the sum of all pi = 1

321
Q

What kind of markov chains have steady state distribution?

A

Regular markov chains

322
Q

What are regular markov chains?

A

Chains where transitoin p^h > 0, that is there are only nonzero entries in the matrix after many transitions.

323
Q

When there is a state i with pii = 1, the markove chain cannot be what?

A

It can not be a regular markov chain. This is called an absorbing state.

324
Q

What is a limiting matrix?

A

A limiting matrix is the transition matrix that is created by P^h as h goes to infinity.

325
Q

What does the limiting matrix look like?

A

It is a matrix where all rows are identical.

326
Q

What is the notation for a population parameter and its estimator?

A
327
Q

What do we need to know to solve uncertainties?

A

we need to know the problems distribution and its parameters.

328
Q

What must we do to gain sufficient information about the parameters of an observed system?

A

collect data

329
Q

What do we use to make statements about a very large set?

A

We use collected and observed samples

330
Q

what is the population?

A

the set that consists of all units of interest.

331
Q

What is a sample?

A

a set of observed units from the population.

332
Q

What is a statistic?

A

any function of a sample. i.e. an arithmetic mean.

333
Q

Although it happens with a low probability, a sample may sometimes give misleading information, a probability of which is?

A

Binomial

334
Q

Sampling and non-sampling errors refer to what?

A

any discrepancy between a collected sample and a whole population.

335
Q

The cause of sampling errors is?

A

the mere fact that only a portion of the population is observed.

336
Q

What causes non-sampling errors?

A

inappropriate sampling schemes or wrong statistical techniques.

337
Q

Three examples of wrong sampling techniques?

A

○ Sampling from a wrong population.
○ Dependent observations- people surveyed together may have opinions dependent on each other.
○ Sampled specimen not being equally likely to be selected.

338
Q

What is simple random sampling?

A

a sampling design where units are collected from the entire population independently of each other, all being equally likely to be sampled.

339
Q

What is an i.i.d?

A

independent identically distributed random variables

340
Q

We consider a sample to be what?

A

a set of random variables obtained by observation.

341
Q

Observations collected by means of simple random sampling design are what?

A

Independent identically distributed random variables.

342
Q

What are simple descriptive statistics?

A

measuring the location, spread, variability and other characteristics that can be computed immediately from a collected sample.

343
Q

What is the mean?

A

measuring the average value of a sample.

344
Q

What is the median?

A

Measuring the central value of a sample.

345
Q

What do quantiles and quartiles show?

A

Where certain portions of a sample are located.

346
Q

Each statistic of a sample estimates what?

A

The corresponding population parameter.

347
Q

What does the sample mean estimate?

How is the sample mean denoted?

What is the definition of sample mean?

A
348
Q

What is expected to converge to E(X) as a sample approaches a larger and larger size?

A

The sample mean.

349
Q

What are the three properties of the sample mean?

A
  1. unbiasedness
  2. consistency
  3. asymptotic normality
350
Q

What is a disadvantage to sample mean?

A

its sensitivity to extreme observations.

351
Q

What is asymptotic normality?

A

the distribution of the normalized estimator converges to standard normal distribution as n approaches infinity.

352
Q

What is consistency, rigorously?

A
353
Q

What does unbiasedness mean?

A

Unbiasedness means that in a long run, collecting a large number of samples and computing the estimator, on the average we hit the population parameter exactly.

354
Q

What is the rigorous definition of unbiasedness?

A
355
Q

This is much less sensitive to extreme observations than sample mean.

A

the median

356
Q

Three Definitions of median, sample median, and population median

A
357
Q

What are the three skewedness of a sample?

A

Symmetric: Median = arithmetic mean

Right-skewed: median < mean

Left-skewed: median > mean

358
Q

How do we find the median of a continuous distribution?

A

We solve F(x) = .5

359
Q

How do we find the median of a discrete distribution?

A
  • For a discrete distribution F(x) = .5 has either a whole interval of roots, in which case any number in this interval excluding the ends is a median, or no roots at all.
  • If there are no roots at all, the smallest x with F(x) >= .5 is the median.
    • It is the value of x where the CD jumps over .5
360
Q

How do you find the median of a sample?

A
361
Q

How can you measure the median speed of cars?

A

Drive so that half of cars overtake you, and half are overtaken

362
Q

the construction of these tell us how well we can expect our sample parameter to match the population parameter

A

Construction of confidence intervals.

363
Q

What is a p-quantile?

A

where p is a percentage

364
Q

Notation for population p-quantile, gamma quantile, quartiles, and medians, and their estimators

A
365
Q

The first, second, and third quartiles are the

A

25th, 50th, and 75th percentiles. They split a population into four equal parts.

366
Q

what is a gamma perentile?

A
  • A gamma-percentile is (0.01gamma) quantile.
367
Q

A median is at the same time what?

A

0.5-quantiles, 50th percentile, and 2nd quartile

368
Q

Notation of how gamma percentiles, quantiles, quartiles and the median relate.

A
369
Q

With a sample of 30, how do we find the .25 quantile?

A
  • For p = 0.25, we find that 25% of our sample of n=30, is np or 0.25*30 = 7.5. For n(1-p) = .75*30 = 22.5. From a sample of 30, we will see that th 8th element has no more than 7.5 observations to the left and no more than 22.5 observations to the right, there fore the 8th element in a sample of 30 is The estimator of Quartile 1.
370
Q

What is sample variance?

A
371
Q

Sample standard deviation measures variability in

A

The same units as x, Variance is in units squared

372
Q

What is standard error?

A

The standard deviation of an estimator

373
Q

Standard errors show

A

precision and reliability of estimators.

  • They show how much estimators of the same parameter can vary if they are computed from different samples.
374
Q

These three estimators are sensitive to outliers, or extreme observations.

A

Sample mean, variance, and standard deviation

375
Q

This is a measure of variability that is not very sensitive to outliers

A

The interquartile range

376
Q

What is the rule for identifying outliers?

A
377
Q

What should we do before we do anything with data?

A

We must look at it

378
Q

IQR = ?

A

IQR = Q3-Q1

379
Q

Six things a quick look at a data graph can suggest

A
  • a probability model, i.e., a family of distributions to be used.
  • Suitable statistical methods for the given data.
  • Presence or absence of outliers.
  • Presence or absence of heterogeneity.
  • Existence of time trends and other patterns.
  • Relation between two or several variables.
380
Q

5 ways to visualize data

A
  • Histograms
  • Stem-and-leaf plots
  • Boxplots
  • Time plots
  • Scatter plots
381
Q

We use hypothesis testing to

A

confirm or reject a statement about a sample population

382
Q

What is the kth population moment?

A

E(Xk)

383
Q

What is the formula for the sample (X1, X2…Xn) kth moment?

A
384
Q

What is the first sample moment?

A

The arithmetic mean xbar

385
Q

What are central moments?

A

Central moments are moments which are computed after the data is centralized by subtracting the mean.

386
Q

For k ≥ 2, what is the kth population central moment?

A
387
Q

the expected value is equal to what?

A

The area above the cdf curve

388
Q

How do we calculate the expected value using only a CDF?

A
389
Q

CDF can find the first moment, but the second moment is found with what?

A

The pmf

390
Q

When are methods of moments used?

A

Under the strong assumption that we know the distribution of our population.

391
Q

How can we find n parameters if we know the distribution of a sample?

A

We take n moments of our sample, and n moments of our distribution and form a system of n equations.

392
Q

When using the method of maximum likelihood, what are we looking for?

A

Assuming we know the distribution, we are trying to find the parameters that would make that sample most likely to be chosen.

393
Q

Method of Maximum likelihood is formally what?

A

maximizing the joint probability of each variable in a given sample according to the distribution. Because they are independent, in the discrete case this is P(x1)*P(x2)…*P(xn)

394
Q

To find the parameters that make the sample the most likely we find where

A

the derivative d/dxP(x) = 0, does not exist, or the boundaries of the sample parameter.

395
Q

How do we simplify maximizing the joint probability of a sample?

A

By using the formula

396
Q

Simplify and use maximum likelihood to find the lambda of a sample that is of poisson distribution

A
397
Q

What are the two methods for estimating parameters of a population?

A

Method of Moments Method of Maximum Likelihood

398
Q

How is the kth population moment defined as?

A
399
Q

What is the kth sample moment defined as?

A
400
Q

What is the first sample moment?

A

The sample mean.

401
Q

What is centralizing the data?

A

Subtracting the mean from each element.

402
Q

What is the kth population central moment defined as?

A
403
Q

What is the second population central moment?

A

Var(X)

404
Q

What is the second sample central moment?

A

The variance, with n replace by n-1

405
Q

How is the k-th sample central moment defined?

A
406
Q

What is the method of maximum likelihood?

A

○ We find such parameters that maximize the probability of getting our data sample.

407
Q

For a discrete distribution, which formula do we maximize?

A

The joint PMF of iid

408
Q

For a continuous distribution, what formula do we maximize to find the data’s parameters?

A

For a continuous distribution, we maximize the joint density.

409
Q

What is a computational shortcut for using the method of maximum likeliness?

A

A nice computational shortcut is to take logarithms first

410
Q

Take the logarithm of the Poisson distribution joint pmf, and maximize it to find the parameter lambda.

A
411
Q

What do standard errors measure?

A

Standard Errors serve as the measures of our estimators accuracy.

412
Q

How is the standard error of our estimated expectation calculated?

A
413
Q

Use the method of maximum likelihood on the exponential density to find Lambda, and use the logarithm shortcut.

A
414
Q

Sometimes the likelihood has no critical points inside its domain, then

A

it is maximized at its boundaries.

415
Q

What is a parameter?

A

A numerical fact about a population.

416
Q

What are all numerical facts about a sample called?

A

Statistics

417
Q

What does lambda hat represent?

A

The rate in a sample.

418
Q

Describe a simple random sample?

A

The sample is drawn from a population randomly, like tickets from a box. Each is drawn without replacement.

419
Q

The size of our sample must not affect what?

A

The distribution of the population, in other words, it must not be so large the distribution of the box no longer holds.

420
Q

What is the expectation and variance of p hat, and the standard error of phat in a binomially distributed sample?

A

E(P) = phat, Var(phat) = p(1-p), SE = sqrt((p*(1-p))/n)

421
Q

What is the standard Error of an estimator?

A

std(estimator)/sqrt(n)

422
Q

SUM Var(Xi) = ?

A

Var Estimator/n

423
Q

What do we need to know to calculate the standard error of a sample?

A

The standard deviation of the population as a whole.

424
Q

What is simple bootstrapping?

A

We decide the sample standard deviation can be used as the population standard deviation.

425
Q

What is the standard error?

A

The standard error is the standard deviation of the expectation.

426
Q

What does unbiased mean?

A

It means that the estimator does not tend to skew to the left or right

427
Q

What is a biased estimator of the Standard deviation of the population, and how do we fix this?

A

The standard deviation of a sample. We divide the Variance of the sample by n-1 instead of n.

428
Q

We are using the standard deviation of a sample as a standard deviation of the population, how do we calculate this?

A

SUM(Xi-X\bar)^2/n-1

429
Q

What is a confidence interval?

A

The range within which the true value resides with some confidence level.

430
Q

When is the confidence interval the narrowest?

A

As n goes to infinity

431
Q

What does a 90% confidence interval mean?

A

It means that if a sampling process was done many times, 90% of the intervals produced would capture the true parameter.

432
Q

Confidence level is not what?

A

The probability that the actual value of the parameter is within the confidence interval.

433
Q

This is often referred to as the confidence level.

A

The coverage probability (1-\alpha).

434
Q

How do we calculate the confidence level using the normal distribution?

A

b = Estimator - Z_alpha/2*Sigma(Estimator)

435
Q

What is the margin of error = ?

A

sigma*Z_alpha/2

436
Q

If we want a confidence interval of 90%, what is alpha and the critical z score?

A

alpha = .1 critical Z score = +/- 1.645

437
Q

Derive the Var(X\bar), and SE or SD(X\bar)

A
438
Q

This is sometimes called the critical probability:

A

1-(alpha/2)

439
Q

When can we use the standard deviation of a sample as the standard deviation of a population?

A

When the sample size is sufficiently large.

440
Q

When must we use to the t distribution?

A

When a sample is normal, but small.

441
Q

We have a small sample from a normally distributed population, C.I. for the mean = ?

A

X\bar +/- (t_alpha/2)*s/sqrt(n) where s is the standard deviation of the sample.

442
Q

What is the degree of freedom?

A

d.f. = v = n-1

443
Q

How do we find t_alpha?

A

T_alpha the anazlog of Z distribution.

444
Q

What happens when the degree of freedom goes to infinity?

A

We get a normal distribution.

445
Q

What is it that we can and cannot tell about a hypothesis?

A

We cannot tell if a hypothesis is true or not; all we can do is determine whether the data provides sufficient evidence against the null hypothesis in favor of the alternative hypothesis.

446
Q

How is the null hypothesis denoted?

A

H_0

447
Q

How is the alternative hypothesis denoted?

A

H_a

448
Q

What is a null hypothesis involving the mean of a population?

A

H_0: our sample mean equals the population mean.

449
Q

What are the three types of alternatives?

A
  1. Two sided alternative: H_a mu does not equal mu. 2. one-sided, left tail H_a: mu is left then sample mu. 3. one-sided, right tail H_a: mu is more than sample mu.
450
Q

Sampling error occurs when we do this:

A

wrongly accept or reject our null hypothesis

451
Q

What does low level of significance mean?

A

it means that only a large amount of evidence can result in rejection of our null hypothesis.

452
Q

What is a type one error?

A

When the result of our test tells us to reject our null hypothesis, although it is true.

453
Q

What is a type II error?

A

When our test tells us to accept our null hypothesis, though it is false?

454
Q

Which type error in hypothesis testing is considered worse?

A

A type II error.

455
Q

What makes a good test result in erroneous decision?

A

Extreme observed data

456
Q

What is the significance level of a test?

A

The probability that the test tells us to reject our null hypothesis wrongly.

CS 3341 (1 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions