This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Stat/Math 390, Winter, Test 2, Feb. 20, 2009; Marzban Open everything, closed messaging/ discussion
Check front page and back page.
Multiplechoice: mark answers on these pages. DO NOT EXPLAIN.
The rest: SHOW your answer and your WORK, on these pages;
Points NO CREDIT FOR CORRECT ANSWER WITHOUT EXPLANATION. e following is a printout from a regression analysis on 14 values of a predictor x and response
y. What is the typical deviation of the response variable from the ﬁt? 5‘: [95970. —1_) Source DF SS MS a) 0.022 .148 c 0.262 d) 2.295 Model I 2.295 2.295
r ‘16:. Error 12 0. .02
— T7, —
2 Thich of the fol owing is true? : \J .D ZZ. <_’/ a) ln the presence of an interaction term in a regression model, R—squared cannot be interpreted in
any meaningful way. b) ln the presence of collinearity, an interaction term must be introduced in the regression model.
c Both a and b. either a nor b. L I (a L L
i  I ur/ 
 L i /. VAX.) 3 . . . 7 2
e samp mg 1str1 utlon of the sample mean 1s glven by Vb] = a /n, 2
a) only when n is large c both a) and b) Laolr ax T51. dZV‘iddlBA
b) only when the population is normal. one of the above. 2 ' A . u n a use we have computed a 95% conﬁdence interval (CI) for the mean lifetime of all lightbulbs. Which of the following is correct? Q a) There is a 5% chance that the population mean is not in our Cl. X
\ .0 7e can be highly conﬁdent that 95% of lightbulb lifetimes lie within the CI. 7\ ﬁThere is a 95% chance that a random sample will yield a sample mean in our Cl. w one of the above. q
@Suppose we have data on x and y, both taking positive values. We look at the scatterplot of log(y) vs. I / I and see a linear relationship. We then ﬁt a regression line and obtain log(y) = 2.0+3.0(I/r).
a) Can we use R2 to assess how well the model does? Why or why not, in one or two sentences? 2 ﬁg. 5’2 assesyeg Tﬁz [await vm‘avxqin £93m via/Mme; b? (i) . b) Do the regression coefficients 2.0 and 3.0 (in some units) have any meaning/ interpretation. If
so, what, in one or two sentences? V9.5. 2.0 : ‘l'ﬁLVaLAR—D’FIJE[~1)Q)¢F53QA as «.500
5 O 7‘ m amtNAT V? baa“?— In 223301) YtS'vJYllrvxg
’f'vbm 1&4an of“? in t 3( £¢5t7BSGMELQ .
ln1ess1on problem y— — oz + ﬁx", we have seen that the sample mean of the ﬁtted values is equal to the sample mean of the y values in the data Show that the sample variance of the
ﬁtted values 1s equal to the sample variance of x values 1n the a es (m2, where ﬂ 1s the OLS estimate of 0. ”4/. Mn 5% FVLCL“ 94‘9““5— 1 gnu'14 +¢5+2 {dark 342=va:)(1—9) \l 41am flawl Naumﬂ 9“,‘45 H
E)
S
I. Ms
E?
1
XI
V
II
J) ‘L (7
0
a
)0 l *—
Simlaur “to 5' 1.
v ‘?' I
3 ‘ s in LossAngaeleslla d makes frequent trips to Washington DC. 60% of the time she travels on airline 1, and the remaining 40% of the time on airline 2. For airline 1, ﬂights are late into
DC for 30% of the time, and late into LA for 10% of the time. For airline 2, these percentages are
25% and 20%. If we learn that on a particular trip she arrived late at DC, what is the probability
that she ﬂew on airline 1? Deﬁne the events clearly, and show the formulas you use, don’t just
add / multiply numbers. Lat 1 2 1n MAT Mm Tm ”tr—ms on'rh'm 1 , 2: 12 .
LA : I: u I, r, ['5 [JL (In—[.0 LA ; 96: (f5 ‘ 17(1) :06: 17(1): 9H J)(Dcll\ = o.) ?(pclz):0.1§ {1.3) (0,4) :%
l7CLAl L) = o .l PCLAl 7.): 9.20 018 M 0.0+
mm) : PCDcll) Pa) _ PM!) 1’1.) @9014) P690) / PCDCIDPOHleznzz) @
?.q8 mu Tngg" “laidfulic who'll; andom samples of Size 1 0° are en from a large area of forest, and the proportion of diseased
trees in each sample is deter mined The actual proportion of diseased trees, 7T, is unknown W hat is
the (lower bound for the) probability that the sample proportion lies within ::0.1 of the population
proportion? Give a numerical answer, and explain your reasoning. (1);7 ) /0‘ / _ —Ol +D__!_. My (770.1<p<'(7+0\) —r«v‘4( 0,? 5—,,
Q!) R" 0" 7:l"7) “bioIF l5 Mod! Ian—LEM '77:: .Zll'. O’T<J_
[ea 20 >FVO‘VC 1110(24 ”'7" ): {>«ob(—o.7,<¥.<o.z)
‘74: ...
View
Full Document
 Spring '08

Click to edit the document details