This preview shows pages 1–3. Sign up to view the full content.
1
STATISTICS FOR MANAGEMENT I
Dec. 9, 1994
19002200 (corrected version)
NAME:
S.N.
Section:
Time: 3 hours
INSTRUCTIONS: ALL ANSWERS (INCLUDING BRIEF EXPLANATIONS) GO ON THE
LEGALSIZED ANSWER SHEET. THIS EXAM QUESTION BOOKLET WILL
NOT
BE
MARKED. However, space is provided here for your rough work. These booklets must be
deposited in the box provided.
Calculators and 1 sheet of notes provided.
Q 1. [4] The welds on tin cans must be done correctly to avoid the possibility of food
becoming contaminated later. Despite a lot of care, about 1 in 8,000 cans is defective. The
bean canning line at H J Heinz receives a shipment of 10 gross of cans (1 gross = 144 = 12
dozen). What is the probability two cans are defective in the shipment.
Q 2. [6] Tinton is a small industrial town with 4000 houses. For a typical day, the average
electrical power requirement of a house is 800 watts with standard deviation 200 watts.
There are also 100 institutional (i.e., government, school or industrial) sites. These require
an average of 16000 watts of power with standard deviation of 4000 watts.
Tinton Hydro has a peak supply capacity of 4.9 Megawatts. If demand exceeds supply,
there will be a brownout or else a complete power cut. If houses use more power in the
evening while institutions use power more during the day, then demand between them
will be negatively correlated. Suppose the correlation is in fact 0.4. What is now the
probability of a brownout or blackout.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Q 3. [3] A street person accosts people to ask for "loose change". Naturally he hopes to get
a Loony, but only 10% of accostees actually give him one. On average, how many people
must he accost before he gets a Loony?
Q 4. The following output comes from an analysis of file sizes in bytes in three directories
on a computer. (Real data!) This information is often important for optimizing the
performance of disk management software. Below we present some numerical and
graphical summaries of this data.
MTB > desc c1c3
N
MEAN
MEDIAN
TRMEAN
STDEV
SEMEAN
cfg
94
7763
208
3719
20809
2146
gmk
17
58795
13147
50802
81030
19653
nis
115
17603
5511
11950
35417
3303
MIN
MAX
Q1
Q3
cfg
3
117001
40
1702
gmk
54
237443
654
103923
nis
207
302728
2645
18974
MTB > let c12=log(c2)
MTB > note log is NATURAL (base e) logarithm = ln
MTB > name c12 'ln_gmk'
MTB > stem c12
Stemandleaf of ln_gmk
N
= 17
Leaf Unit = 0.10
1
3 9
1
4
2
5 7
6
6 3448
6
7
8
8 23
(1)
9 4
8
10 06
6
11 146
3
12 023
a) [2] Compute the interquartile range for the files in the directory "cfg".
b) [2] Compute the coefficient of variation for files in the directory "nis".
c) [2] Compute the 40th percentile of the log of the length of files in the directory "gmk".
d) [1] From your answer in (c), what is the 40th percentile of the length of files in the
directory "gmk"?
e) [2] Using the descriptive statistics, what set of file sizes has the greatest dispersion?
This is the end of the preview. Sign up
to
access the rest of the document.
 Spring '00
 Phansalker
 Management

Click to edit the document details