[toc]

## Ch 2

### Population

$\mu: \text{population mean}$
$\begin{align}
\mu=\frac{\sum{x_i}}{N}
\end{align}$

$\sigma: \text{populatoin variance}$
$\begin{align}
\sigma&=\frac{\sum(x_i-\mu)^2}{N}\\
&=\frac{\sum x_i^2}{N}-\mu^2
\end{align}$

### Sample

$\bar x:\text{sample mean}$
$\begin{align}
\bar x =\frac{\sum{x_i}}{n}
\end{align}$

$s^2: \text{sample variance}$
$\begin{align}
s^2&=\frac{\sum(x_i-\bar x)^2}{n-1}\\
&=\frac{\sum x_i^2-\frac{(\sum x_i)^2}{n}}{n-1}
\end{align}$

$s: \text{sample standard deviation}$
$\begin{align}
s=\sqrt{s^2}
\end{align}$

### Stem and Leaf Diagram

**e.g. 1**


**e.g. 2**

### Histograms

**e.g. 1**

**e.g. 2**

### Bar chart

**e.g. 1**

### Pareto chart

Basically bar graph but sorted.

**e.g. 1**

### Box Plots


**whisker**

the min/max point within 1.5 IQR from the quartiles

**outliers**

points within 3 IQR

### Sample Correlation Coefficient

$\begin{align}
S_{xy}&=\sum(x_i-\bar x)(y_i-\bar y)\\
&=\sum x_i y_i -\frac1n(\sum x_i)(\sum y_i)
\end{align}$

$\begin{align}
r=\frac{S_{xy}}{\sqrt{S_{xx}S_{yy}}}
\end{align}$

## Ch 3

### Random variable

a variable whose measured value can change in a random experiment.

**e.g. 1**

toss the coins, the total number of heads observed is called a random variable.

### Discrete (Countable)

countable object, finite or infinite

### Continuous

infinite and uncountable, measurable

### Probability Density Function (PDF)

$P(a<X<b)=\int^b_af(x)dx$


if X is continuous random variable

$P(x=a)=0$

### Cumulative Distribution Function (CDF)

$F(x)=P(X\leq x)=\int^x_{-\infty}f(u)du$

### Mean (Expected value)

$\mu=E(X)=\int^\infty_{-\infty}xf(x)dx$

### Variance

$\sigma^2=V(X)=\int^\infty_{-\infty}(x-\mu)^2f(x)dx=E(X^2)-\mu^2$

The **standard deviation** of $X$ is $\sigma$

### Normal distribution (Gaussian distribution)

Bell-shaped and symmetrical

denote as $X\sim N(\mu,\sigma^2)$

### Standard normal random variable

$\begin{align}
Z=\frac{X-\mu}{\sigma}
\end{align}$

$P(X\leq x)=P(Z\leq z)$

where $Z$ is a standard normal random variable, and $z=(x-\mu)/\sigma$ is the z-value obtained by standardize $x$.

### Normal Probability Plots

For determining whether sample data conform to a hypothesized distribution.

### Probability Mass Function

for **discrete** random variable

$f(x_i)=P(X=x_i)$

### Cumulative Distribution Function

for **discrete** random variable

$F(x)=P(X\leq x)=\sum f(x_i)$

### Mean and Variance

the same except the integral is replaced by sum.

### Binomial Distribution

A trail with only **two possible outcomes** is call **Bernoulli trial**.

usually assumed
1. the random experiment are **independent**.
2. the probability of a success on each trial is **constant**

#### Mean & Variance

$\mu = E(X)=np$

$\sigma^2=V(X)=npq=np(1-p)$

### Poisson Process

Number of events that occur in an interval.

### Poisson Distribution

#### Mean & Variance

$\mu=\sigma^2=\lambda=np$

$\begin{align}
p(x)=\frac{\lambda^xe^{-\lambda}}{x!}
\end{align}$

### Exponential Distribution

Is a continuous distribution.

**e.g.**
The length of time or the distance between occurrences of random events.

$\begin{align}
\theta = \frac{1}{\lambda}
\end{align}$

#### pdf

$f(x)=\lambda e^{-\lambda x}$

#### Mean & Variance

$\begin{align}
E(X)=\frac{1}{\lambda}
\end{align}$

$\begin{align}
V(X)=\frac{1}{\lambda^2}
\end{align}$

Mean = SD = $\theta$

### Normal Approximation to Binomial and Poisson Distributions

1. Requires large sample size
2. Need correction for continuity
$np\geq 5 \& nq\geq 5$

### formulas


### Power

$1-\beta$ (type II error)

The power of a statistical test is the probability of rejecting the null hypothesis $H_0$ when the alternative hypothesis is true.

### P-Values in Hypothesis Testing

the smallest level of significance that would lead to rejection of null Hypothesis $H_0$

### Large Sample Test

if $n\geq 30$, the sample variance $s^2$ will be close to $\sigma^2$ for most samples

### Confidence Interval

95% of our confidence intervals will contain $\mu$ and 5% will not.

#### confidence level (confidence coefficient)

$1-\alpha$ (type I error)