1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
|
# Notes for third probability lecture
*Moment generation function* is a third way to describe the behavior of a random variable.
Can also find *Key performance indicators*.
Which will be explained in this document.
## Expectation
Is just the mean value.
$$
E[X] = \sum_{i} x_i P(X = x_i) = \sum_{i} x_i p(x_i)
$$
In the continues way integral is used instead.
$$
E[X] = \int_{\infty}^{\infty} x f(x) \mathrm{dx}
$$
Can also calculate expectation distribution function, however this can only be used if all values are non-negative:
$$
E[X] = \sum_{k=0}^{\infty} P(X > k) \\
E[X] = \int_{0}^{\infty} (1 - F(x)) \mathrm{dx}
$$
Tatianas recommendation is to calculate expectation from the PDF.
### LOTUS
### Some properties
$$
E[a X + b] = a E[X] + b
$$
The mean of a constant is the constant itself:
$$
E[b] = b
$$
### Multiple variables
If $Z = g(X,Y)$ one can find the expectation with:
$$
E[Z] = \sum_{i} \sum_{j} g(x_i, y_j) \cdot p(x_i, y_j)
$$
If continues just use integrals instead.
$$
E[Z] = \int_{-\infty}^{\infty} \int_{-\infty}^{\infty} g(x, y) \cdot f(x, y) \mathrm{dxdy}
$$
The following rule can be used:
$$
E[X+Y] = E[X] + E[Y]
$$
If $X$ and $Y$ are **independent** the following is true:
\begin{align*}
E[g_1(X) \cdot g_2(Y)] &= E[g_1(X)] \cdot E[g_2(Y)] \\
E[X \cdot Y] &= E[X] \cdot E[Y]
\end{align*}
## Variance
Describes the mean of the distance between outcomes and the overall mean.
Good way to describe the spread of the random variable.
\begin{align*}
Var(X) &= E[(X - E[X])^2] \\
Var(X) &= E[X^2] - E[X]^2
\end{align*}
If there is no power of two, it will be mean minus mean, which wont work.
One can define the *standard deviation* to bring back the unit from squared.
$$
Std(X) = \sqrt{ Var(X) }
$$
A rule for variance:
$$
Var(a X + b) = a^2 Var(X)
$$
The variance of a constant is therefore $0$.
### Summing
$$
Var(X+Y) = Var(X) + Var(Y) + 2 Cov(X, Y)
$$
If X and Y are independent the Cov part disappears.
## Covariance
\begin{align*}
Cov(X,Y) &= E[(X - E[X]) \cdot (Y - E[Y])] \\
Cov(X,Y) &= E[XY] - E[X] \cdot E[Y]
\end{align*}
Shows whether two variables vary together, can be both positive and negative.
If it is possible $X$ and $Y$ are varying from the average together.
Some rules below:
\begin{align*}
Cov(X, X) &= Var(X) \\
Cov(a X, Y) &= a Cov(X, Y) \\
Cov(X + Y, Z) &= Cov(X, Z) + Cov(Y, Z)
\end{align*}
If X and Y are independent, then covariance is zero (X and Y are *uncorrelated*).
X and Y can be uncorrelated and not be independent.
## Correlation coefficient
It is hard to compare covariance as the value is dependent on the size of X and Y values.
We can therefore take the Correlation coefficient instead.
$$
Corr(X,Y) = \frac {Cov(X,Y)} {\sqrt{Var(X)} \cdot \sqrt{Var(Y)}}
$$
vim: spell spelllang=da,en
|
