Basic Statistics - (06) Probability Distribution
1. Random Variable(X)
Variables whose possible values are numerical outcomes of a random phenomenon.
- Discrete
- countable number of distinct values
- Continuous
- infinite number of possible values
- probability distribution
- Discrete
- probability mass function
- Continuous
- probability density function
- a certain internal under the curve rather than the height of the curve.
- probability density function
- Discrete
2. Cumulative Probability Distribution
-
CPD answers two questions
-
Quantile
3. The Mean of a Random Variable
- \[\mu_x\]
-
mean of random variable
-
add or subtract two random variables: X+/- Y
\[\mu_{a+bX}=a + b*X\] \[\mu_{x+y} = \mu_x + \mu_y\]
4. Variance of a Random Variable
-
variance 方差
方差用于衡量随机变量或一组数据的离散程度,方差在在统计描述和概率分布中有不同的定义和计算公式。
-
variance equations
- plus and minus
-
Two Variables X and Y
\[var(X + Y) = var(X) + var(Y) {\color{BurntOrange}+} 2cov(X,Y) \\ var(X-Y) = var(X) + var(Y) {\color{BurntOrange}+} 2cov(X,Y)\]
- covariance 协方差
\(cov(X,Y) = E{[(X-\mu_x)(Y-\mu_y)]}\)
公式简单翻译一下是:如果有X,Y两个变量,每个时刻的“X值与其均值之差”乘以“Y值与其均值之差”得到一个乘积,再对这每时刻的乘积求和并求出均值(其实是求“期望”,但就不引申太多新概念了,简单认为就是求均值了)。
5. Normal Probability Distribution 正态分布
-
表达式
-
公式
\[f(x) = {1\over\sqrt{2\pi}\sigma} * e^{-{1\over2}*({x-\mu\over\sigma})^2}\]-
μ:总体期望均值
-
σ^2:方差
-
x:随机变量
-
Z score
- \[{x-\mu}\over{\sigma}\]
-
-
μ& σ
6.Empirical Rule(68-95-99.7 法则)
7. Z-Score
-
z-score
-
standardized data
-
How to calculate Z-score?
8. Binominal Distribution 二项分布
-
binomial distribution=discrete probability distribution
-
公式
-
属性
\[\mu = n * p\] \[\sigma = \sqrt{np(1-p)}\]n: trials
p: probability of success