This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: DALHOUSIE UNIVERSITY FACULTY OF SCIENCE
Department of Mathematics and Statistics
STATISTICS 2080 Midterm Examination Date and Time: 6:00 pm — 7:30 pm, March 2, 2011 Name: Student ID #: Signature: The entire exam is marked out of 50 points (and worth 25% of your ﬁnal grade). It is closed
book, but you may refer to a formula sheet, and you may use a calculator. The number of
points allocated to each portion of a problem is shown in the margin. To get maximum credit,
show all your work. (6) 1. Nine people undertook a weight loss program. Before the program their mean weight was
88 kilograms, and after the program it was 85 kilograms. The sample standard deviation
computed for the weight difference was 4 kilograms. Is there sufficient evidence to con
clude that people lose weight by undertaking the program? Carry out the appropriate
hypothesis test at signiﬁcance level 0.05, showing all your steps. ~)’i(3:/~Ad‘1CD Vg Via=ud>o (“wq §=a $24 J )
{obs :— 1 2:. é“ a erg
5/“ 47/3 C(2me “i9 ﬁle) ér‘xg'imlaoixcn P“\Iaiu€. between 6221.025 0rd 0.03.
“*5 (836.6% HQ «:3— ol‘: 0.03 Or ‘Yke weirﬁwi' \o§'§ program work}. (7) (4) Fees “= "3787 F 2. (3.) Suppose we want to compare the means of 5 groups, and we have 9 observations for each group. Complete the following ANOVA table. GarWWW . .  ,  _ I .—
...uhr . . .n. .. .‘r _‘.>... on ll
— 
 I In. v. .' rg—__.un.. 01S Q3/\<§aap§ , (V:— 4‘5 obsmud‘xmﬁ
U512 3'ng “l QSE =3 (b) Carry out a hypothesis test to determine if the group means are the same (use a signiﬁcance level of a = 0.05) Ho; Mﬁiuz" M3 :M4: M3
Vs Ha  A; $14K ‘Yor Some ink /‘ RNOVA
Cl.“me F454”) A.\§'°§(\lout\m 0d" d=0.0S t: ——> r2393: H0
(means QAQ diﬁwnﬁ b5 > ma: 00§ 3 (6) (c) The sample means for groups 1 and 2 were 14.5 and 13.8 respectively. With the
overall signiﬁcance level of the ANOVA in part (a) being a = 0.05, use the Bonfer—
roni correction and determine Whether the means of group 1 and 2 are signiﬁcantly different from one another. 04 9533 0(‘4003, F=(:)=\O —‘5 OL¥: «:2 ‘0 =.DOS Tar—)4" i402 M‘:M.Ll \Ig, 7 M‘ﬁﬂg ‘32}: \4.§ J 32;:— R58
‘0‘: q J (‘1: q “in. +0“ = Y“ *2 _ Masst M \ — ..
Mama" tate‘egei’z “ 300$? Compwt. "(u ‘bN_Q “:2: t4@ Hugo Sxd‘m’l‘) J MQE ‘' re 6 d"
ae =‘>
Pva‘wa 4 .OOl less ’ikcm 5%— = 00025 Ho
/‘ q 390
' (5) ((1) Determine a Bonferroni conﬁdence interval for the mean difference group 1 and group 2. Again recall the overall signiﬁcance level of the ANOVA in part (a) is a = 0.05).
Are the means different from one another? (Note: when looking up in the t—table use the closest value). a
l o . a( 0‘ _. _ ,l xvi; Mew/“Wurst; (gr5Q)
or: t (2ﬂ‘7t)(0.\%‘1‘5)
on t 0.463
=75 (0.1345) Luise) “Re C1 does heft anctivdﬁ zero —9 (“eded H0
(grmp 1. anti. ’2. Moms 0A2, olgﬁeﬂgﬂ) 4 (8) 3. The water temperature at 3 separate sites in Kejimkujik Park was recorded. The data is
given below 1 10 9 2 14 11 8
7
6 Test the hypothesis Ho : p1 = #2 = #3 using the KruskalWaﬂjs test (use a = 0.01). Rank mbgeAvdeﬂ‘
Sﬁe \ S‘Hre. '2. SH; :3 ’1 S 4
c1 Q» 3
\o 8 f
R = 8M eras 2m
{"1 1' 3 '5 4
0g
T€S+ §+atK§AQC z u M . ﬂ 2‘
K" with?) Z; ULRRL 7.) t2 :2.
—‘ z:  a: v UH \\ .
“(n+0 (tow!) "08‘ 6‘ 1 “=5; = 5: a 9g IL " $
"9 K: .\OC1051§3§‘8.E1“Sg)*‘ Hera8.33
~ a» Mtg—$91} = “2.43
“a.
Compw ‘3‘“) (XL d‘ﬁr‘bﬁl‘m a 4 do ain’t" Wk. €33“ H I 1 l i 4. Shorebirds numbers were surveyed at two different habitat types: a/mudﬂat and a beach.
i The following counts were obtained: 25 13
’ 14 16
15 18 17 20
19 9 (8) (a) Using a pooled two sample t—test, determine Whether these data provide evidence
that Mudﬂat has more birds than the Beach (use a signiﬁcance level a = 0.05). iszes A, 'SMs. 4.08m ) QM:
Vagigfl ) $8 31%,?”14 J mg: (“$3 u; + 4 (4122.4 3 2: [’L ggss; (“j—wake, 10.1
H—5 reeled“ HQ no em mm W‘s. w 16:. i
hear) b u“ «24% S. as», 6 (6) (b) Carry out the same hypothesis test as in part (3) using non—parametric Wilcoxon
Rank Sum test. What do you conclude? {261 AM: “*ﬂe <¢:\:\ aﬁa
’ mud‘haf Beach M Z
’3 g.
4— ‘7
e 51
a .1
to M 5%.)?“ 6g: N91 ; AWN' QM (NH31’2. = gin.) x“; 2 3g, erw‘l— .. {\Mng(w+\3/\1: (9)6192) a: 3 \2. O m} 51w = 3.47“? TQS’C Qicaizgﬁcu‘: '2ng z: ‘  ~ I .= LOCK/1 ...
View
Full
Document
This note was uploaded on 03/27/2012 for the course ECON 2280 taught by Professor Daniel during the Spring '12 term at Dalhousie.
 Spring '12
 daniel

Click to edit the document details