Undergraduate
คณิตศาสตร์

[Stat I] final

253

7056

4

fuqinaz

fuqinaz

Stat I ภาค IE ปี 2 เทอม 1 คณะ วิศวฯ จุฬา สรุปไว้ตอนไฟนอล
ตอนนั้นเพิ่งหัดใช้แอพเลยเละหน่อยนะ 55

PromotionBanner

ノートテキスト

ページ1:

JOINT
PROB
from textbook.
Outline
- Joint Prob. (a marginal, PDF, PMF)
มี independence ว
-Expected Value
- Covariance
-
- Correlation coefficient

ページ2:

5-17,2RV
Joint pmf: denoted as fag (x,y); X,Y (only) octive range
ใช้อธิบาย ข้อมูล discrete ที่มี R > 2 เช่น ในตาราง, ซึ่ง สมบัติ คือ
Number of Bars of Signal Strength
y Response time
(nearest second) 1
2
3
4 0.15
0.1
0.05
3
0.02
0.1
0.05
20.02
0.03
0.2
10.01
| 0.02
0.25
Joint pdf; denated as
Lay (x, y)
สมบัติ
เช่น ox เป็นหนอง ()
(1) Fxy (xy)>0
Fxy (x,y) = P(X=x, Y=y)
(พวกเรนจ์ X กับ Y จะมี prob. เป็น 0)
ใช้อธิบายข้อมูล continueur ที่ R22, prob คือ 1. ให้กราฟ assume ว่านอกกราฟ พท. 20
A joint probability density function for the continuous random variables X and Y,
denoted as fxy(x, y), satisfies the following properties:
(1) fxr(x, y)20 for all x, y
fxy(x, y)
(2)
]] fx (x, y) dx dy = 1
ด้านนอก =
(3) For any region R of two-dimensional space,
P((X,Y) = R)= [fxx (x, y) dx dy
marginal: von prob. You RV onis w jaint
7.80 7.70 7.60
เช่น ถ้าจะบอก margins ทอง X3 6) P(X=39 ) มารวมกัน
fx (3) P(X-3)= P(X=3, Y=1)+P(X=3, Y=2)+P(X=3,Y=3)+P(X=3,Y=4)
-0.25 +0.2+0.05+0.05 0.55
3.05
3.0
2.95
ถ้าทา margin แต่ละ X กัน ในตารางของ ฯ ข้างบนจะได้มัน MEDE[Y] BY
x=Number of Bars of Signal Strength
Marginal
Probability
Distribution of Y
EX] 1(0.2)+2(0.25)+3(0.55)
E[Y]=4(0.3)+3(17) +2 (-25) + 1 (.28)
y Response time
(nearest second) 1
2
3
40.15
0.1
0.05
0.3
Fy (4)
30.02
0.1
0.05
0.17
2 0.02
0.03
0.2
0.25
10.01
0.02
0.25
0.28
0.2 x (1)
0.25 J (2)
0.55 (3)
5762/26 continuous Timos integrate -
→
Marginal Probability Distribution of X
แบบนี้ทำได้ แต่ discrete
fx (x)=fxx(x, y)dy and fr (y)=√fxy(x, y) dx
สำหรับ X กรรม ทุก ๆ
y dx

ページ3:

Mazha Joint & marginal
Example 5-2
Server Access Time Let the random variable X denote the time until a computer server con-
nects to your machine (in milliseconds), and let Y denote the time until the server authorizes you
as a valid user (in milliseconds). Each of these random variables measures the wait from a common starting time and
X< Y. Assume that the joint probability density function for X and Y is
fxr (x,y) = 6 X 10 exp(-0.001.x – 0.002y) for x<y
ตอนแรกต้องตั้งให้เป็นก่อน ขอบอกแค่ xy -
ต้องเริ่มที่ 3 แล้วไปถึง
ถึง 4 .. ไม่ปอกอมมน
6x1
-0.001x-0.0024 dydx g³ ³
ง
คือตรงที่แรง)
สมมติถ้าอยากได้แค่ที่ 2000, X1000 ก็จะ เอาแต่น
Downดูจะได้ pres.
P(X<1000, Y<2000)
$1000/2000
=
6x10e
dydx
2000
zumes. marginal
Example 5-4
0 1000
Server Access Time For the random Variables that denote times in Example 5-2, calculate the
probability that Y exceeds 2000 milliseconds.
This probability is determined as the integral of f (x,y) over the darkly shaded region in Fig. 5-7. The region is
partitioned into two parts and different limits of integration are determined for each part.
กลับมาดูรูปจะพบว่าท้อง แขกช่วง
Improb vy-
ง 2000
P(Y>2000)
marginal y
จะ
- ช่วงนี้ น
2000
ช่วงนี้ไม้ 72002
2000
2
ท่าปกติจะได้แบบนี้
2000
co
P(4) 2000) = 2000 dydx + 1,000 dydx
ได้
nd in marginal y Fo
100
2000√x
ry-1dx
-
จะได้สมการรวด
you
..
√20

ページ4:

Conditional prob. distri
พอเพิ่ม condition ทั่วไป ค่า prob. ก็จะเปลี่ยน เช่นจากตารางนี้
x = Number of Bars of Signal Strength
y = Response time
(nearest second) 1
2
3
40.15
0.1
0.05
30.02
0.1
0.05
20.02
0.03
0.2
160.00
0.02
0.25
จะเป็นว่า ถ้า x-3 จะมีโอกาส 1
มากกว่า X-1
P(9=1|X=3) > P(Y-1) X=1)
ในเมื่อไม่เท่ากัน แสดงว่า R X เป็น dependent IV คือ prob. ตัวนึงจะขึ้นกับอีกตัวด้วย
mn midterm.. PLAIB) = - เอามาใช้ต่อเลย โดยในจะใช้ 9 ทน A, X B
PLANB)
PLB)
P(=1,X=3) _ 0.25.
0.05+0.05 +0.2+0.25
จริงๆแก่เอามาคิดว่าเป็นสัดส่วนเท่าไหร่ของการนั้น
จะได้ P(Y=1)x=3) = P(X=3)
เรียกว่า y given Xx
สําหรีม Continuous
Given continuous random variables X and Y with joint probability density
function fxy(x, y), the conditional probability density function of Y given
X = x is
Ji, (y) = -
fxx (x, y)
fx(x)
for Jx (x)>0
(5-4)
อย่าลิม
Because the conditional probability density function fn (y) is a probability density
function for all y in R,, the following properties are satisfied:
(1) fru (y)=0
(2) J. (y)dy=1
(3) P(Y € B | X = x)= [fr. (y)dy for any set B in the range of Y (5-5)

ページ5:

หวอยม
Example 5-2
Server Access Time Let the random variable X denote the time until a computer server con-
nects to your machine (in milliseconds), and let Y denote the time until the server authorizes you
as a valid user (in milliseconds). Each of these random variables measures the wait from a common starting time and
X<Y. Assume that the joint probability density function for X and Y is
ARA
fxy (x, y)=6×10 exp(-0.001x-0.002y) for x<y
Tum conditional pdf for Y given X=x
→ marginal x when x>0 ñ'D26
→
Yo
- fx 20 0.003e"
-0.003%
face) = f = dy-
for 70
Fx4 (x4)
0.002X-00024
for x70 and xcy
• fyx (y)= Fu(x) 0.0036-0.003x = 0.902 e
→JUNG= P(4>2000 | X=1500)
- X คงที่ ก็แทน 500 ลงใน K
- 107472000 @dy
นี่เลย จะได้ 0.002%
0.002 (1500)-0.002y
↳ P (4) 2000/X=1500) = 0.002 2004 dy
e
-0.0024∞
=0.002e3 [4]
-0.002 2000
=
Ans
ที่ทำก็เหมือนเปลี่ยนตรงบนหล่ม
x=Number of Bars of Signal Strength
201
y Response time
(nearest second) 1
2
3
4 0.15
0.1
0.05
3 0.02
0.1
0.05
20.02
0.03
0.2
10.01
0.02
0.25
33210.8
Y
given X=x
y Response time
x = Number of Bars of Signal Strength
แต่ละตัวจะ เป็น prob. ในแต่ละครลัมน์
(nearest second)
1
2
3
4 | 0.750
0.400
| 0.091
30.100
0.400
0.091
x=Number of
20.100
0.120
0.364
10.050
0.080
0.454
1
Total
1
1
1
0.15
0.750
0.10070.2
0.100
0.050
1

ページ6:

Independence
บางที ก็ไม่เปลี่ยนตาม - ใช้ margins จะง่ายกว
xerzim: fxy (x,y) = fxcxifyly) → ₤xx (y) =
for a xay
1xxfam
= fyry) → F4ixy = fyys
Fxx)
(4) P(XEA, YEB) = P(XEA) P(YB) for any sets
For random variables X and Y, if any one of the following properties is true, the
others are also true, and X and Y are independent.
(1) fxr (x, y)=fx(x) fy (y) for all x and y
(2) fr (y) fr (y) for all x and y with fx (x)>0
(3) fx (x)=fx(x) for all x and y with fy (y)>0
A and B in the range of X and Y, respectively.
ที่เหลือใ
ใช้ กอัน
independent
(5-7)
Rectangular Range for (X,Y)
มีไว้ check คร่าวๆ ว่าเป็น independent ใหม่ ถ้า range ไม่เป็น) คือ ไม่ใช่ independent
In D set of points in two-dimensional
Space
2000
0
n
dependent
D 19/0
Th
Cheek propaties
independent]
0
2000
คือสมมติขอแบบนี้ คือไม่เป็น in... 109000%
เพราะช่วงที่ hiligh ไม่เป็น 0

ページ7:

Example 5-11 Independent Random Variables Suppose that Example 5-2 is modified so that the joint
probability density function of X and Y is far (x, y) = 2×10 *exp(−0.001x-0.002y) for x≥ 0 and
y≥0. Show that X and Y are independent and determine P(X>1000, Y<1000).
Note that the range of positive probability is rectangular so that independence is possible but not yet demonstrated.
The marginal probability density function of X is
fx(x)=√2×1000
0.001-0002 dy=0.001c -0.001x
for x>0
The marginal probability density function of Y is
คุณกันได้
fr (y)=2×10-001-0.002; dx = 0.002 e¯
for y>0
Therefore, f(x, y) f(x) f(y) for all x and y, and X and Y are independent.
To determine the probability requested, property (4) of Equation 5-7 can be applied along with the fact that each
random variable has an exponential distribution. Therefore,
Independent
107 probrum P(X>1000,Y<1000)=P(X>1000) P(Y<1000)
= '(1-e)-0.318
108 ปี คงไม่ออกไม่มีในสไลด์ ช่างแม่ง ๆ

ページ8:

Corerience : มอก กระจายตัว 4. ค. สัมพันธ์ของ RV > 2 ตัว
Fon covariance Most Expected Value
→
• Expected Value You for 2 RVs
E[h(X,Y)]=(xy) for(x, y)
ffh(x, y) fxx (x, y) dx dy
9 น
X, Y discrete
(5-13)
X, Y continuous
[x] weighted average you hoxy) via=paus range (X, Y) You
- บอกค่าที่พอลองสุ่ม สัก ๑๐ ครั้งจะได้เท่านี้
Covariance Oxy=E[(X-MX) (Y-My)]; Mx = E[X], My = E[Y]
| = 5.00 5.2 (x - Mx)ly - My) fxy (x,y) dx dy
Covariance Oxy = EXYJ ECXJELY]
มันนอกค.สัมพันธ์ยังไง? ... ลอง assume ว่าแต่ละจุด (qually, likely
(X-Mx) (4-My)
ถ้าเป็นบวก แสดงว่าถ้า X
เพิ่ม y เพิ่ม
Cov Niwan linear relationship 1167 Cov == 1/2/12/ linear
Cov+
ค. น+
ถ้าเป็นลมคอตรงข้าม
อันนึงเพิ่มอันนึงลด
(a) Positive covariance
Cov
ด.ชH-
All points are of
equal probability
(b) Zero covariance
5
Oxy 20
62/2/ "uncorrelated
๔ (จะไม่มอก
relationship
(e) Negative covariance
X,Y independent
(d) Zero covariance
-> Cov=Q
1105 Cov=o X, Y = Yai independent 1/

ページ9:

Properties of Covariance
.
Cov (X, X)
.
Cov (X, Y)
Var(X)
Cov(Y, X)
• Cov (cX, Y) = c Cov(X, Y)
.
Cov (X, Y+Z) Cov(X, Y) + Cov(X, Z)
Var(X+Y) Var(X) + Var(Y) +2 Cov(X, Y)
Var[X]
=
ΣΣCov(X1, X₁]
Var[X]+2 Cov[Xi, Xj]
If X's are independent
-
-
Cov[X, X]=0, i zj.
So, Var(X) =Σ₁ Var(X)
i j>i

ページ10:

Correlation is "Correlation coefficient" nia "normallize covariance">
บอกคงสัมพันธ์ของ RVs ได้เหมือน Covenance แต่ง่ายกว่า
The correlation between random variables X and Y, denoted as pxy, is
Pxy=
cov(X, Y) OXY
=
√V(X)V (Y) σx Gy
(5-15)
ยังไง 661
ปืน +
-I < Pxy < 1
ง
Pxy
มีเครื่องหมาย +
+ - ม Oxy
→
+
เงิน
ใช้ที่มาก, สัมพันธ์ linear ระหว่าง 2PM ที่มีหน่วยต่างกันได้
ถ้า 2x, 40 - เรียกว่า " Correlated” วัน
Pxy =
Com(X,y) = (
= 0
- D1 2RVs & independent is
แต่ถ้า Corn (X,y) = 0 ไม่ได้แปลว่าจะเป็น independent
ถ้า 2 - 21 ก.สัมพันธ์จะเป็นเส้นตรง y = a +6 ; 240 ลบ
ไม่ได้แปลว่า X₁Y จะเพิ่ม จด พร้อมกัน
- P
9กล้ 1 ไม่
กราฟแนวโน้มเป็นแบม C
<on linear relationship)

ページ11:

Esti
mation

ページ12:

Point Estimation
เราจะอธิบายลักษะ อง
11m2/non parameter.
pap
D= unknown
Estimator (e) 12/2 RV & 22/07.
Jordan
•μ,6 for quantitive pap.
{^ for binomial pap.
parameter
statistic arvasin
1. Bint Estimation : คำนวณay ตัวนึงมา estimate 9
2. Interval Estimation: PINY 2 Men 12/21Your estimate &
x²² ^ up; p = x
• estimator YOUMN: Sample Yaiming (smal sampling dist.)
↳ unbiased to mean VAN estimator = parameter (E[^] =>
0
ê
☑
=; X~ Binomial (hp)
M
P
02
S²
SECX-X)²
n-1
n-1
• ! (Ex² - (Ex)}); V(Y) - E(Y) - [E(Y)]²
ECY-V(Y)+[E(Y)]²
Property of Paint Estimator
-
- prefer dist. Var ugo
3.4. บอกช่วงที่ว่าจริงจะ เปียงไปจาก 6 ได้
- WD n 201→
• 69/26 normal
- 95% YON point estimate 7: 2025 126 1.96 sd
L 1 sd 12/26 unknown parameter - 95 standard error You estimator
Standard error an Estimator <
A estimate sd YAN estimator)
-ใช้ในการ estimate 4 ว่าดูเหมือนจะต่าง ๆ แค่ไหน
- SD = ~ unknown 6 → Estimated sd of X: [sd= 1/1/
- binomial: estimator p = ✗/: sd = √1
The Margin of Error
Pl-p)
n
(d/o unknown
P
4/ d = (1-5)
L
1std error of
the estimation
n
119721
Margin of e
error = 1.96x
95%
quant. (n>30)
\
bina (np25,925)
£1.96 S
√n
+1.96
True value
n

ページ13:

Interval Estimstion
fairly
Confidence Interval
- สร้างช่วงที่ค่อนข้างจะชัวร์ ว่าข้อมูลจริงอยู่ในช่วงนี้
sure no 90%
(a: significant coefficient
2 confidence
coefficient, 1-27 0.9
95%
ส่วนใหญ่จะใช้ 0.20, 0.95 0.99
ไม่รู้จะใช้ไรก็ใช้ 0.2% ไม่
2.25
•parameter.
L ± 0.96SE
Estimator
- ถือซะว่า pop, กระจายแบบ normal และไม่รู้ 6 - 7 2 จาก 2
- เนื่องจากเราไม่รู้ว่าจริงของ Petrometer
พอเอาหลายๆ ช่วงมาเรียนจะได้ประมาณนี้
X-M
6/55
P(-1.96 (1.96) = 0.95
Gr√n
-1.966 <X-M<1.965 In
-x-1.96<<-x+1.96
PC-1.96x+1.96% n ) = 0.95
มีอันนึงไม่มี แอยู่ในช่วง - เป็น 5. error
"
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Interval number
Estimator £1.965E

ページ14:

- การเปลี่ยน Confidence level
- เปลี่ยนค่า ที่ใช้ตา
ตามตาราง 2-3
Tail area Z1-a/2
.05
1.645
90%
1-0a
.025
1.96
95%
.005
2.58
99%
Zde Estimator Zı-diz
100 (1) CI :
Estimator ZSE (σ =SE)
-Large-Sample CI
for quantitative pop
for binomial
pop
100 (1-x)% CI for mean u
ตัวอย่าง
100(1-2)% CI for propotion p
pq
L random sample of
n=50m26 THAN 756 g/day wl sd = 35 g
find a 95% CI for pop. average μ.
Saln
35
12.
→756 ± 1.96
→756 ± 7.90
√50
Find a 99% confidence interval for u, the population
average daily intake of dairy products for men.
d
930 746.30 << 765.90
+
Sal x±ave 756 ± 2.58
35
→
→756₤12.77
√√50
หรือ 743.23 << 766.77 4
Of a random sample of n = 150 college students, 104 of <
the students said that they had played on a soccer team
during their K-12 years. Estimate the proportion of
college students who played soccer in their youth with a
98% confidence interval.
Sal =104
=.69
150
n
2012
= 20.01
- 2.33
.. .69± 8.33
(69) (.31)
150
→.69±.09 or 60 <p<.78

ページ15:

One-sided confidence bound.
บอกแค่ด้านเดียวก็ได้ ทำได้ 2 แบบ
9. 9 ปบนเน
Yoหลื
Supper-confidence bond: usx+zao m
lower-confidence bond: x-zan u
Choosing the Sample size
n
Jun. x estimate u, error E-/X-MID= 5 Zove th
1-1-2121√
E = error Fu
ไม่พ้น หอกไป .. E (2
ที่เหมาะสมที่ 1-4 จะไม่เกิน ที่กำหนดค
24 E (สมมติให้ อยู่รอบเลย)
* (***) ( ถ้า ๆ ไม่มี ให้ ปัดขึ้น
h=
<
ถ้า binomial ใช้
(ในสไลด์ใช้ 2 SE 48 พอได้ค่า .. ก็ อามานได้เท่ากัน
E
E=B= margin of error
or width of CI
Z242
√ ₁ =E; p = q = .5
Student's t distribution
177 sample size 157 22/256 11:55:51 + Youth's normal 11: dio degree
of freedom,n-1
t = X-A
s/m
ทางจะยาวกว่า ที่ยกว่า norms xp n มากจะใกล้ arms (อง
Graphs of t density functions
t1-a.k-ta,k
0
ta.k
t

ページ16:

ดูค่า 6 วัน 4 ค่าที่ได้จะเป็นค่า critical ทอง
31.821
For a random sample of size n =
6965 10, find a value of t that cuts off
3747 .025 in the right tail.
3.143 Row= df=n−1 = 9
df
Case
Can
1
3,078
6.314
12.706
1.886
2.920
4.303
3
1.638
2.353
3.182
4.541
1.533
2.132
2.776
5
1476
2015
2.571
3.365
6
1440
1.943
2.447
1
1415
1.895
2.365
2.998
1397
1860
2.306
2.896
9
1.383
1833
22624
To
1372
1.812
2.228
2.764
11
1.363
1.796
2.201
2.7
12
1.356
1.782
2.179
2.681
7.025 = 2.262
13
1.350
1.771
2.160
2.650
14
1.345
1.761
2.145
2.624
15
1.341
1.753
2.131
2.602
2.821 Column subscript = a = .025
46
46
Inference of small sample for u
สําหรับ 100 (1-4)2.CI ของ
Two-sided: X± tove, n√
Sampling Distr. of sample Var.
upper confidence bond: ± tx,n-1 3/1
normal sample in var=5² 4200
chi-squared x²= (n-1) s² (×;-)²
62
=
mean df, variance = 2df
ธเ
pop var. 61
Chi-square w/ df=n-1
เปาวา แต่พอ n มากจะใกล้ normal
f(x) 4
| P(x"> x²+) =α
(ตอนการใช้ค่า C. กับ 4 - h-1 ดู
f(x)
Lower 5%
upper 5%
0
(a)
Xak
0.05
0
X0.95.10
= 3.94
(b)
ที่แรงงาน พท.
X₁.05, 10
= 18.31
= α
อย่าลืมว่า เป็นกราฟไม้ ถ้าผืน To-sided ต้องดูค่าทั้งบน-ล่างเพราะ ไม่เท่ากัน
0.05

ページ17:

Hypothe
sis
testing

ページ18:

Single Sample Hypothesis Test
1
สนใจแค่ 1, p. 6 โดยจะมีข้อมูลอยู่แล้ว แต่จะเอามาเชื่อว่า ควรจะ reject ข้อมูลนั้นไทย (จะไม่
ยอมรับ เพราะไม่รู้ค่าที่แท้จริง และจะทา Sample size 1 ที่ยอมรัน error ได้
ส่วนใหญ่จะใช้กับบ เช่น มี 4 จาก process
Introduction
Joy
ละ
สาวอีก
I process in claim s
2 แบบว่าคนนั้น
มันจะคล้ายๆ กันในศาลเวลามีผู้ต้องสงสัยตามา ซึ่งจะเป็นไปได้ 2
1. เป็นผู้กระทำผิด (quilts)
2. เป็นผู้บริสุทธิ์ (innocent) Assume ว่าเป็น innocent ก่อนแล้วค่อยหาหลักฐานม) reject
ส่วนประกอบของ hypothesis testing
1. The null hypothesis, Ho
innocent (guilty)
เป็นสมมติฐานแรกที่ assume ว่าจนกว่าจะมีหลักฐานพอจะ reject
2. The alternative hypothesis, Hy
จะถูก acept ว่าทุกเมื่อ aprove H,
อยากพิสูจน์อะไรให้ทั้งที่ 4
เช่น อยากรู้ว่าทำผิดไหม : 1. :guilty
Note - สมมติฐานต้องอ้างถึงข้อมูลของ Pop
เท่านั้น
-
1 กับ 2 ต้องต่างกันโดยพันเอง เช่น H. : 45 46 4 4 4
ผมเปรียบเทียบ : ในศาล
Ho innocent
Aiguilty
8. test statistic is p-value
✓
วันนี้
A : ไม่ถึงจนที่ claim
H: N. claim
เฟนถ่ายยาก ทดสอบกา
1 μ x 12/26 test statistic, p-value no prob Ho
เม่นจริง
4. Rejection region : You/การปฏิเสธ สมมติฐาน : ขึ้นกับ 6 ที่ใช้ ปกติใช้ 01 ไม่ก็ .05)
NN
YOU-a test statistic nio p-value minumns= reject
5. Conclusion
2 2 112121 1. Reject Ho 2. Do not reject Ho

ページ19:

ตัวอย่าง
The mayor of a small city claims that the average income in his
city is $35,000 with a standard deviation of $5000. We take
a sample of 64 families, and find that their average income
is $30,000. Is his claim correct?
ว่าพูดจริงก่อน แล้วทุกหลักฐานมา rejec
- assume ว่าอันนี้จริงก่อน
มาทดลองว่า mayor พูดจริงไหม โดย assume
ม่
my Ho: M= 35,000 (mayor worse).
Ha: μ # 35,000 (mayor
mintons, test statistic = 30,000
CLT X~N (35,000
Reject<
442
5000
164
=-30000-35000
5000
√64
n=64 >309 CLT
.8
→
Mon Fomisos
หนวดนั้น
Z212
'
1+α
acceptance
region
ตาม สมมติ ครอมเป็น 2-2
1. Critical val.
2. p-value
=
21%2
-
2
> Yako bi iximiz → [ reject / mullistava rejection
(ปกติจะ 1-2 ปี)
d
region
reject uns rejection region and
มี 2 วิธี แต่ละวิธีต้องได้ผลเหมือนกัน)
1. Critical value approach
ถ้า x 24, หรือ 24/2
reject
Reject
+12
( แม่ลงทอม region ปืน 2)
2. p-value approach
951. →α=0.05 Z₁-02 = 1.96 <2 reject
Plz (-2) x2 = 0.04996 < 0.05 reject
reject Ho ND Zail prob <α
(แปลง 2 ที่ได้มาเป็น prob
เทียนกัน 4 ง่ายๆ
- ถ้าเปลี่ยน 4 ไปๆมาๆ อันนี้เรียบง่าย) - 4, ต้องเป็น 2 บนม

ページ20:

ino 4 rejection region
ถ้าค่า ไม่ไปไกลนาดนั้น จะดูยังไง? - ๔/๑๗
2
reject
→
reject
ฝ่า
2/2
219
Acceptance 21-12
region
21. Critical value Approach: reject Ho if Z test stat > Z1-0/2
L2. p-value approach (12/22 Zctest stat) 11/16 prob
reject
(แปลงาน region เป็น2)
or Ztest stat <Z21
reject
reject H. if
α-1
Acceptance
Z region
- ปลายทางอยู่ช่วงนี้ - reject
ตาม สมมติได้คอมเป็น 2-0
1.
Critical val
->
Ztail prob <
.. p-value=α. ก็ไม่ reject
จะเรียกว่า critical value
TANG 95%. α = 0.05 Z₁-012 = 1.96 <2 reject
2. p-value
=
= P(Z<-2) x2 = = 0.04996 < 0.05 reject
- ถ้าเปลี่ยน 4 ใน มาๆ อันนี้เที่ยง่ายกว) 44-45 ต้องทำเป็น 2 ผมป

Comment

fuqinaz
Author fuqinaz

Gootnotes ครับ

ผู้เยี่ยมชม

พี่ครับรบกวนถามหน่อยครับ ใช้อะไรจดอ่าครับ

fuqinaz
Author fuqinaz

thx ka preaw

Kittpraew
Kittpraew

love your notes!