Basic Data Science : Statistical Reviews

Mean

\(\large \mu = \sum_{i=1}^N(p_iN) = E(X) \)

Rule of mean

R 1 : เมื่อ  C = constant,  \( E(C)  = C \)

R 2 : เมื่อ  C = constant,  \( E(X+C)  = E(X) + C  \)

R 3 : เมื่อ  C = constant,  \( E(CX)  = C \times E(X)   \)

R4 :  \( E(X+Y) = E(X) + E(Y)\)


Variance

\( \large \sigma^{2}_{x} = \sum_{i=1}^{N}p_i[X_i - E(X)]^2 = VAR(X) \)

หรือ

\( \large \sigma^{2}_{x} = E[(X_i - E(X))^2]\)

หรือ

\( \large \sigma^{2}_{x} = E(X^2) - (E(X))^2\)


Rule:

R1 : \( Var(C) = 0\)

R2 : \( Var(X+c) = E[X_{i}+c - E(X+c)]^2 = Var(X)  \)

R3 : \( Var(cX) = c^2Var(X) \)

R4: \( Var(X+Y) = Var(X) + 2COV(X,Y) + Var(Y) \)

R5: \(\large \sigma^2 = \frac{\sum(x-\bar{x})^2}{n} \)

\(\large n\sigma^2 = \sum(x^2-2x\bar{x}+\bar{x}^2) \)

\(\large n\sigma^2 = \sum{x^2}-2\bar{x}\sum{x}+\sum{\bar{x}^2} \)

\(\large n\sigma^2 = \sum{x^2}-2n\bar{x}^2+n\bar{x}^2,\sum{x}=n\bar{x} \)

\(\large n\sigma^2 = \sum{x^2}-n\bar{x}^2 \)

\(\large \sum{x^2} = n\sigma^2 +n\bar{x}^2 \)

\(\large \sum{x^2} = n(\sigma^2 +\bar{x}^2)\dashrightarrow*** \)






















ความคิดเห็น