Docsity
Docsity

Prepare for your exams
Prepare for your exams

Study with the several resources on Docsity


Earn points to download
Earn points to download

Earn points by helping other students or get them with a premium plan


Guidelines and tips
Guidelines and tips

Hypothesis Testing: A Comprehensive Guide with Examples, Study notes of Statistics

Sampling distribution, Point estimates, Interval estimation, Hypothesis testing, Analysis of Variance

Typology: Study notes

2020/2021

Uploaded on 03/13/2021

176114
176114 🇬🇧

4

(1)

5 documents

1 / 7

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
p
-
value
,
significance
level
.
Type
2
Type
I
error
.
Power
Oneftu
.
-
sided
tests
for
normal
mean
t
-
test
with
unknown
6
'
and
n
Test
for
normal
variances
Test
for
comparison
of
2
normal
means
with
known
variances
or
unknown
,
equal
variances
Test
for
correlation
coefficients
Tes t
for
ratio
of
normal
variances
pf3
pf4
pf5

Partial preview of the text

Download Hypothesis Testing: A Comprehensive Guide with Examples and more Study notes Statistics in PDF only on Docsity!

p

  • (^) value , significance^ level^.^ Type^

Type I^ error^.^ Power ① (^) Oneftu.^ -^ sided^ tests^ for normal^ mean ② t -^ test (^) with (^) unknown 6 ' and n ⑤ (^) Test (^) for normal variances ④ Test^ for (^) comparison of^ (^2) normal means with known^ variances ⑤ or unknown^ , (^) equal variances Test (^) for correlation (^) coefficients

⑥ Test for ratio of normal variances

Chapter

9 Hypothesis Testing O (^) 9. tp-va P- value (^) is the probability of^ the event that the^ test (^) statistic takes the^ observed

value or more extreme values under Ho^.

Small ' p

  • value ' Ho is not supported (^) by the^

data

large '

p

  • value' Ho is (^) not inconsistent with (^) the data Significance level =^ a^ B the^ probability of rejecting a^ true Ho kepeotttoifp-value.ee# O (^) 9. 5. Ttests^ for^ normal^ means Ho '^. N = μ.^ Hi: (^) U #^ Mo l Xi^ -^ -^ - , Xnl (^) for (^) NCN (^) , 62 ) 62 known F- Us Under Ho^ , In NC (^) Me, E (^) ) ¥

n N lo^. I )

i (^) : (^) Test (^) statistic T = TITI n = t M¥1:^ calculate^ p -^ value Pu. (^ Text^ )^ = pl 121 >

  • t (^) ) = p - (^) value compare p
  • (^) value with significance level^ a Method find^ Critical^ value^ Ca such that^ PCH>^ Ca^ ) =^ a p
  • value at it and^ only if ITI^ >^ Cd I Reject Ho O 9.5.40 tests^ for normal^ means Ho :^ N -^ -^ Mo^ Mi :^ Ms^ Mo Under Ho.^ test^ statistic^ T -^ -^ Tntx- up n (^) Nco. 1) For a^ given 10007. significance level D= (^) Pu. ( (^) TEC ) = plz SC^ )

under no^ ,^ =÷E÷^ n Reject Ho it t^ >^ Want for (^) any 6276.2^ , power of^

the test :

PC

  1. = Ps C Ho^ rejected ) = Ps IT >^ X'd,^ n^ -^ t^ ) =P, ( (^) H >^ x'ant^ ) = Ps CHIT

> ÷

'

. x'a.m ) > a as^ - I probability of^ Type I^ error^ I^ -^ PCG

  • As^6 ' increases , (^) power function (^) pc 6 )^ increases
  • (^) For (^) H, : (^) 621= Reject He (^) tf t^

' X'

I - E. n - I or^ -17 (^) Ago (^) , n - I ⑧ o 9.lt#amanmUpaHdbW~~yi~N( (^) Me

( Xi , Yi ) Xin NC^ Mx, 6 × 2 ) , Gi)

independent Ho :^ Nx -^ - he let Zi^ -^ -^ Xi^ - Yo (^) , Zi -^ N ( (^) Nx- Me ,^6 ×^465 ) IEEE T=¥=5h% Reject Ho (^) Nx- up when^ I-471% (^) ,^ n^ -^ I^ if th^ :^ Mx#^ My t >^ tan-^ I^ it Hit Ux>^ My

t c^ - taint if Hi : Mx^ any

(^0) Power function

For Ho '^ -^ Nx^ - Ny H^ , :^ Mx>My

μ= Nx-^ My^70 play

    • Pa ( Ho repeated) = pm ( (^) T - ta int ) =pnF¥ >^ tent^ )

=pn( This^ >^ tan (^) ,^ - I )

°

Tha

Ash (^) increases → g- increases →^ power increases 0 9. Gmparmgtnonorma1mean€ of^ different (^) sample sizes { Xi (^) ,^.^ -^.. Xn^ ) f Yi^ ,^.^.^ - , Yml^ mtn

I- a' E. Xi F- II. Yi

sie (^) # (^) E. Hi -^ H^ ' Sy

. - this Hi-FI^ ' F (^) , -7 (^) , Sj , Sy ' are (^) independent

In Nlnx, ) HY n x'n^ -^ i

E. nwcuy^ ,^ I^ cm-"f -^ him^ -^ I F- (^) Tn N tax- my (^) , #t (^) ) If (^6) ×2= l-cnx-ny,yj TZ

  • heon, 7- μs×Y6× . tlm-YSYY6.ie//ntm-2 X'mtn - z

= x }T-(nI

~ Enema

Ymt Yn^ CH) sit (^) cm- HST ① With known^65 by '

Ho : Nx - My Hi : Mx # My

Under (^) Ho (^) , Mx- Ny - T

    • T Jtym n Nco^ ,^ I^ ) lool (^) I- a) %^ Confidence^ interval^ for^ Nx- Ny

I

T I Z%×6xYym

H=lplJT increases as (^) Ifl increases 09.14 (^) -estsfrtheratiooftnenormalvaria## ( Xi^.^.^ -^ -^ -^ Xml^ f^ Yi^ ,^.^ -^ - ,^ Yml^ from^ thnx.^6 ×^4 N^ (Ny (^) , 65 ) Independent Ho (^) : (^) f÷=k Hi : %÷tk We (^) have cn% n (^) x'ni , ' n x'mi ( n-115×^2 -6× 2 :-(htt^ Y6× 5 ×^2 i÷÷÷÷÷÷÷→ Reject Ho (^) if t -^ Fi- E. n - i. (^) my or^ t >^ FE ,^ n^

  • I (^) , m - I

1004-41 confidence interval^ for^ {÷ :

( (^) FI-E.n-i.mx (^) S¥ , FE.n-i.mu (^) ×S¥ (^) )

  • To^ find lett^ tail^ of F

FI- d (^) Nish = Fa , vav, E. (^) g.

Fo -0515.3 =

To find F 3. s cleft tail)

PC Fs^ .sc^ fI , = #=^ o (^) - N (^) ) =