Very large in magnitude and others are small, then the geometric mean is a better

representative of the data than the simple average. In a "geometric series", th

e most meaningful average is the geometric mean (G). The arithmetic mean is very
biased toward the larger numbers in the series. An Application: Suppose sales o
f a certain item increase to 110% in the first year and to 150% of that in the s
econd year. For simplicity, assume you sold 100 items initially. Then the number
sold in the first year is 110 and the number sold in the second is 150% x 110 =
165. The arithmetic average of 110% and 150% is 130% so that we would incorrect
ly estimate that the number sold in the first year is 130 and the number in the
second year is 169. The geometric mean of 110% and 150% is G = (1.65)1/2 so that
we would correctly estimate that we would sell 100 (G)2 = 165 items in the seco
nd year. (D.) Median:
Median: The median is the middle value in an ordered array of observations. If t
here is an even number of observations in the array, the median is the average o
f the two middle numbers. If there is an odd number of data in the array, the me
dian is the middle number. The median is often used to summarize the distributio
n of an outcome. If the distribution is skewed, the median and the interquartile
range (IQR) may be

better than other measures to indicate where the observed data are concentrated.
Generally, the median provides a better measure of location than the mean when
there are some extremely large or small observations; i.e., when the data are sk
ewed to the right or to the left. For this reason, median income is used as the
measure of location for the U.S. household income. Note that if the median is le
ss than the mean, the data set is skewed to the right. If the median is greater
than the mean, the data set is skewed to the left. For normal population, the sa
mple median is distributed normally with m = the mean, and standard error of the
median (p/2) times standard error of the mean. The mean has two distinct advant
ages over the median. It is more stable, and one can compute the mean based of t
wo samples by combining the two means. (D.) Mode: The mode is the most frequentl
y occurring value in a set of observations. Why use the mode? The classic exampl
e is the shirt/shoe manufacturer who wants to decide what sizes to introduce. Da
ta may have two modes. In this case, we say the data are bimodal, and sets of ob
servations with more than two modes are referred to as multimodal. Note that the
mode is not a helpful measure of location, because there can be more than one m
ode or even no mode. When the mean and the median are known, it is possible to e
stimate the mode for the unimodal distribution using the other two averages as f
ollows: Mode 3(median) - 2(mean) This estimate is applicable to both grouped and
ungrouped data sets.
Question 5: Explain the classical approach to the probability theory. Also expla
in the limitation of classical definition of probability.
Answer:
The classical approach to probability is to count the number of favorable outcom
es, the number of total outcomes (outcomes are assumed to be mutually exclusive
and equiprobable), and express the probability as a ratio of these two numbers.
Here, "favorable" refers not to any subjective value given to the outcomes, but
is rather the classical terminology used to indicate that an outcome belongs to
a given event of interest. What is meant by this will be made clear by an exampl
e, and formalized with the introduction of axiomatic probability theory.

Classical definition of probability If the number of outcomes belonging to an ev

ent E is NE, and the total number of outcomes is N, then the probability of even
t E is defined as.
Limitation of classical definition of probability
There are basically four types of probabilities, each with its limitations. None
of these approaches to probability is wrong, per se, but some are more useful o
r more general than others. In everyday speech, we express our beliefs about lik
elihoods of events using the same terminology as in probability theory. Often, t
his has nothing to do with any formal definition of probability, rather it is an
intuitive idea guided by our experience, and in some cases statistics. Probabil
ity can also be expressed in vague terms. For example, someone might say it will
probably rain tomorrow. This is subjective, but implies that the speaker believ
es the probability is greater than 50%. Subjective probabilities have been exten
sively studied, especially with regards to gambling and securities markets. Whil
e this type of probability is important, it is not the subject of this book. A g
ood reference is "Degrees of Belief" By Steven Vick (2002). There are two standa
rd approaches to conceptually interpreting probabilities. The first is known as
the long run (or the relative frequency approach) and the subjective belief (or
confidence approach). In the Frequency Theory of Probability, probability is the
limit of the relative frequency with which an event occurs in repeated trials (
note that trials must be independent). Frequentists talk about probabilities onl
y when dealing with experiments that are random and well-defined. The probabilit
y of a random event denotes the relative frequency of occurrence of an experimen
t's outcome, when repeating the experiment. Frequentists consider probability to
be the relative frequency "in the long run" of outcomes. Physical probabilities
, which are also called objective or frequency probabilities, are associated wit
h random physical systems such as roulette wheels, rolling dice and radioactive
atoms. In such systems, a given type of event (such as the dice yielding a six)
tends to occur at a persistent rate, or 'relative frequency', in a long run of t
rials. Physical probabilities either explain, or are invoked to explain, these s
table frequencies. Thus talk about physical probability makes sense only when de
aling with well defined random experiments. The two main kinds of theory of phys
ical probability are frequentist accounts (such as Venn) and propensity accounts
. Relative frequencies are always between 0% (the event essentially never happen
s) and 100% (the event essentially always happens), so in this theory as well, p
robabilities are between 0% and 100%. According to the Frequency Theory of Proba
bility, what it means to say that "the probability that A occurs is p%" is that
if

you repeat the experiment over and over again, independently and under essential
ly identical conditions, the percentage of the time that A occurs will converge
to p. For example, under the Frequency Theory, to say that the chance that a coi
n lands heads is 50% means that if you toss the coin over and over again, indepe
ndently, the ratio of the number of times the coin lands heads to the total numb
er of tosses approaches a limiting value of 50% as the number of tosses grows. B
ecause the ratio of heads to tosses is always between 0% and 100%, when the prob
ability exists it must be between 0% and 100%. In the Subjective Theory of Proba
bility, probability measures the speaker's "degree of belief" that the event wil
l occur, on a scale of 0% (complete disbelief that the event will happen) to 100
% (certainty that the event will happen). According to the Subjective Theory, wh
at it means for me to say that "the probability that A occurs is 2/3" is that I
believe that A will happen twice as strongly as I believe that A will not happen
. The Subjective Theory is particularly useful in assigning meaning to the proba
bility of events that in principle can occur only once. For example, how might o
ne assign meaning to a statement like "there is a 25% chance of an earthquake on
the San Andreas fault with magnitude 8 or larger before 2050?" (See Freedman an
d Stark, 2003, for more discussion of theories of probability and their applicat
ion to earthquakes.) It is very hard to use either the Theory of Equally Likely
Outcomes or the Frequency Theory to make sense of the assertion.

QUANTITATIVE TECHNIQUES IN MANAGEMENT

Assignment B Question 1: Write a note on decision making in management. How one
will take decision under risk and uncertainty.
Answer:
Decision-making is a crucial part of good business. The question then is how is a
good decision made? One part of the answer is good information, and experience
in interpreting information. Consultation ie seeking the views and expertise of
other people also helps, as does the ability to admit one was wrong and change o
nes mind. There are also aids to decision-making, various techniques which help t
o make information clearer and better analysed, and to add numerical and objecti
ve precision to decision-making (where appropriate) to reduce the amount of subj
ectivity. Managers can be trained to make better decisions. They also need a sup
portive environment where they wont be unfairly criticised for making wrong decis
ions (as we all do sometimes) and will receive proper support from their colleag
ue and superiors. A climate of criticism and fear stifles risktaking and creativ
ity; managers will respond by playing it safe to minimise the risk of criticism wh
ich diminishes the business effectiveness in responding to market changes. It may
also mean managers spend too much time trying to pass the blame around rather t
han getting on with running the business. Decision-making increasingly happens a
t all levels of a business. The Board of Directors may make the grand strategic
decisions about investment and direction of future growth, and managers may make
the more tactical decisions about how their own department may contribute most
effectively to the overall business objectives. But quite ordinary employees are
increasingly expected to make decisions about the conduct of their own tasks, r
esponses to customers and improvements to business practice. This needs careful
recruitment and selection, good training, and enlightened management. Types of B
usiness Decisions 1. Programmed Decisions These are standard decisions which alw
ays follow the same routine. As such, they can be written down into a series of
fixed steps which anyone can follow. They could even be written as computer prog
ram

2. Non-Programmed Decisions. These are non-standard and non-routine. Each decisi

on is not quite the same as any previous decision. 3. Strategic Decisions. These
affect the long-term direction of the business eg whether to take over Company
A or Company B 4. Tactical Decisions. These are medium-term decisions about how
to implement strategy eg what kind of marketing to have, or how many extra staff
to recruit 5. Operational Decisions. These are short-term decisions (also calle
d administrative decisions) about how to implement the tactics eg which firm to
use to make deliveries. Figure 1: Levels of Decision-Making
Figure 2: The Decision-Making Process

The model in Figure 2 above is a normative model, because it illustrates how a g

ood decision ought to be made. Business Studies also uses positive models which
simply aim to illustrate how decisions are, in fact, made in businesses without
commenting on whether they are good or bad. Linear programming models help to ex
plore maximising or minimising constraints eg one can program a computer with in
formation that establishes parameters for minimising costs subject to certain si
tuations and information about those situations. Spread-sheets are widely used f
or what if simulations. A very large spread-sheet can be used to hold all the know
n information about, say, pricing and the effects of pricing on profits. The dif
ferent pricing assumptions can be fed into the spread-sheet modelling different pr
icing strategies. This is a lot quicker and an awful lot cheaper than actually c
hanging prices to see what happens. On the other hand, a spread-sheet is only as
good as the information put into it and no spread-sheet can fully reflect the r
eal world. But it is very useful management information to know what might happe
n to profits what if a skimming strategy, or a penetration strategy were used for
pricing. The computer does not take decisions; managers do. But it helps manager
s to have quick and reliable quantitative information about the business as it i
s and the business as it might be in different sets of circumstances. There is,
however, a lot of research into expert systems which aim to replicate the way real
people (doctors, lawyers, managers, and the like) take decisions. The aim is th
at computers can, one day, take decisions, or at least programmed decisions (see
above). For example, an expedition could carry an expert medical system on a la
p-top to deal with any medical emergencies even though the nearest doctor is tho
usands of miles away. Already it is

possible, in the US, to put a credit card into a hole-in-the-wall machine and get
basic legal advice about basic and standard legal problems. Constraints on Decis
ion-Making Internal Constraints These are constraints that come from within the
business itself. - Availability of finance. Certain decisions will be rejected b
ecause they cost too much - Existing Business Policy. It is not always practical
to re-write business policy to accommodate one decision - Peoples abilities and
feelings. A decision cannot be taken if it assumes higher skills than employees
actually have, or if the decision is so unpopular no-one will work properly on i
t. External Constraints These come from the business environment outside the bus
iness. - National & EU legislation - Competitors behaviour, and their likely resp
onse to decisions your business makes - Lack of technology - Economic climate Qu
ality of Decision-Making Some managers and businesses make better decisions than
others. Good decision-making comes from:1. Training of managers in decision-mak
ing skills. See Developing Managers 2. Good information in the first place. 3. M
anagement skills in analysing information and handling its shortcomings. 4. Expe
rience and natural ability in decision-making.

5. Risk and attitudes to risk. 6. Human factors. People are people. Emotional re
sponses come before rational responses, and it is very difficult to get people t
o make rational decisions about things they feel very strongly about. Rivalries
and vested interests also come into it. People simply take different views on th
e same facts, and people also simply make mistakes.
Question 2: The Mumbai Cricket Club, a professional club for the cricketers, has
the player who led the league in batting average for many years. Over the past
ten years, Amod Kambali has achieved a mean batting average of 54.50 runs with a
standard deviation of 5.5 runs. This year Amod played 25 matches and achieved a
n average of 48.80 runs only. Amod is negotiating his contract with the club for
the next year, and the salary he will be able to obtain is highly dependent upo
n his ability to convince the teams owner that his batting average this year was
not significantly worse than in the previous years. The selection committee of t
he club is willing to use a 0.01 significance level. You are required to find ou
t whether Amods salary will be cut next year. Answer:
Null Hyopothesis -Ho: Amods batting average this year (48.80) is not significantl
y different from his all-time batting average of 54.50 Alternative Hypothesis -H
a: Amods batting average this year (48.80) is significantly lower than his all-ti
me batting average of 54.50 = 0.01
t=
48.80 - 54.50 5.5 / 25
= -5.1818
The critic l v lue of t is -2.492 t df = 24 Conclusion: Reject Ho nd ccept H
(Amods b tting ver ge this ye r is signific ntly lower th n his ll-time b ttin
g ver ge. Amods s l ry will most likely be cut next ye r.

3 The s l ries p id to the m n gers of comp ny h d me n of Rs. 20,000 with

st nd rd devi tion of Rs 3,000, Wh t will be the me n nd st nd rd devi tion if
ll the s l ries re incre sed by 1) 10% 2) 10% of existing me n 3) Which polic
y would you recommend if the m n gement does not w nt to h ve incre sed disp rit
ies of w ges? Answer
1) 10% Both the me n nd st nd rd devi tion will simply incre se by 10% to Rs 22
,000 nd Rs 3,300, respectively. 2) 10% of existing me n Only the me n will incr
e se by 10% to Rs 22,000 nd the st nd rd devi tion will rem in the s me t Rs 3
,000. 3) Which policy would you recommend if the m n gement does not w nt to h v
e incre sed disp rities of w ges? Incre sing the s l ries by 10% of existing me
n does not incre se disp rities of w ges, therefore, is recommended.

C se study
Ple se re d the c se study given below nd nswer questions given t the end. Ku
sh l Aror , second ye r MBA student, is doing
study of comp nies going publi
c for the first time. He is curious to see whether or not there is signific nt
rel tionship between the sizes of the offering (in crores of rupees) nd the pr
ice per sh re fter the issue. The d t
re given below: Size (in 108 39 68.40 5
1 10.40 4.40 crore of rupees) 12 13 19 12 6.50 4 Price ( in rupees) Question You
re required to c lcul te the coefficient of correl tion for the bove d t set
nd comment wh t conclusion Kush l should dr w from the s mple. Answer: N 1 2 3
4 5 6 TOTALS X 12 13 19 12 6.5 4 66.5 Y 108 39 68.4 51 10.4 4.4 281.2 XY 1296 5
07 1299.6 612 67.6 17.6 3799.8 X 144 169 361 144 42.25 16 876.25
2
Y 11664 1521 4678.56 2601 108.16 19.36 20592.08
2
r=
6(3799.8) - (66.5)(281.2)
[6(876.25) - (66.5) ][6(20592.08) - (281.2)
2
2
= 0.67
Conclusion: There is

positive correl tion for the bove set of d t

Assignment C (Objective Questions)

Answer ll questions. Tick M rk () the most ppropri te nswer. 1. Which of the f
ollowing is not correct bout construction of b r ch rts? . All b rs should ris
e from the s me b se line
b. Width of the b r should be proportion l to the d t represented
c. The b rs should be rr nged from the left to right d. Length of the b rs shou
ld be proportion l to the d t represented 2. Which of the following is not true
bout me n bsolute devi tion :. Me n devi tion is obt ined by c lcul ting the bsolute devi tions of e ch obs
erv tion from me n
b. Me n devi tion is more comprehensive me sure comp red to r nge c. It is con
ducive to further lgebr ic tre tment d. It c nnot be computed for distributions
with open end cl sses. 3. The v lue index number me sures the- . Ch nges in pri
ces of b sket of commodities from one period to ncre sed (stogr m : ) The cl
ss interv ls re represented by the b se of the rect ngles b) The frequencies r
e represented by the heights of the rect ngle c) If the cl ss interv ls re of e
qu l width then the b ses of the rect ngles will be equ l in length
d) The t llest rect ngle in histogr m represents the cl ss interv l with the l
owest frequency
10. Which of the following me sures in not ffected by the presence of extreme v
lues in d t set : ) R nge b) Arithmetic me n c) St nd rd devi tion
d) Medi n
11. 1\2x + 1/3y 1/3z = -1 1/3x 1/2y -1/6z = 4 1/6x 5/6y + 1/2z = 3 The v lues of
x in the bove simult neous equ tions would be- ) 3
b) 6
c) 9 d) 12
2
12. The following det ils re v il ble with reg rd to d t set: Sx = 33, Sx =
199, n = 6. If e ch observ tion in the d t set is multiplied by 2 then the st
nd rd devi tion of the resulting v lues will be equ l to:
) (35/3)1/2
b) 35/3 c) 3 d) 25 13. The following d t pert ins to three commodities:Commodit
y Rice Whe t Pulses Price in 2004 (Rs. /kg) 11.50 13.50 26 Price in 1994 (Rs. /k
g) 9.50 8.50 20

The b se ye r is 1994. The unweighted ggreg tes price index for the ye r 2004 i
s pproxim tely-14. If the regression equ tion is the perfect estim tor of the dependent v ri bl
e then which of the following is f lse? ) The st nd rd error of estim te is zer
o
b) The coefficient of correl tion is zero
c) The coefficient of determin tion is 1.00 d) All the d t points f ll on the r
egression line 15. If the regression equ tion is perfect estim tor of the depe
ndent v ri ble then which of the following is f lse : ) The st nd rd error of es
tim te is zero
b) The coefficient of correl tion is zero
c) The coefficient of determin tion is 1.00 d) All the d t points f ll on the r
egression line 16. Which of the following represents the proportion of v ri tion
in the dependent v ri ble th t is expl ined by the regression line :) Coefficient
b) Coefficient
stim te 17. If
ween -1 nd 0,

of determin tion
of correl tion c) Coefficient of v ri tion d) St nd rd error of e
the coefficient of correl tion between the two v ri bles lies bet
then the cov ri nce between them is- ) Positive

b) Neg tive
c) Zero d) Equ l in m gnitude to the v ri nces of both the v ri bles 18. If bYX
is the slope of coefficient of regression line of Y on X, nd bXY is the slope c
oefficient of regression line of X on Y then which of the following is true : )
bYX is positive implies th t bXY is positive
b) bYX is positive implies th t bXY is neg tive
c) bYX nd bXY re reciproc ls d) The product of bYX nd bXY is zero 19. A gr ph
ic l method of representing st tes of n ture nd courses of ction involved in d
ecision m king is referred to s-) Decision tree
b) Histogr m c) Sc tter di gr m d) Frequency distribution 20. If the prob bility
of occurrence of one event is not ffected by the occurrence of nother event
nd vice vers then the two events re s id to be- ) Collectively exh ustive
b) Independent
c) Dependent d) Mutu lly exclusive

21. B yes theorem helps the st tistici n to c lcul te-- ) Dispersion b) Subjectiv
e prob bility
c) Posterior prob bility
d) Cl ssic l prob bility 22. In binomi l distribution the prob bility of getti
ng zero or more numbers of successes is equ l to- ) 0
b) 1
c) The prob bility of getting zero success d) The prob bility of getting success
es in ll tri ls 23. Which of the following me sures represent the sc tter of th
e v lues in d t set : ) Arithmetic me n b) Geometric me n
c) St nd rd devi tion
d) Medi n 24. As the s mple size incre ses- ) The v ri tion of the s mple me n f
rom the popul tion me n becomes l rger
b) The v ri tion of the s mple me n from the popul tion me n becomes sm ller
c) The v ri nce of the s mple becomes less th n the v ri nce of the popul tion d
) The st nd rd devi tion of the s mple becomes more th n the st nd rd devi tion
of the popul tion. 25. In the gr phic l method of solving line r progr mming pro
blems if there is unique optim l solution, then the optim l solution- ) Is lw
ys found t the center of the fe sible region b) Is lw ys t the origin c) Lie
s outside the fe sible region
d) Is loc ted t one of the corner points of the fe sible region
26. A multiple regression equ tion h s- ) Multiple dependent v ri bles b) One in
dependent v ri ble
c) One dependent v ri ble
d) A st nd rd error of estim te equ l to zero
27. Which of the following conditions indic te the existence of multiple optim l
solutions when line r progr mming problem is solved by the gr phic l method :
) One of the constr ints is p r llel to the horizont l xis b) The objective fu
nction is p r llel to the vertic l xis
c) The objective function is p r llel to one of the edges of the fe sible region
which is in the direction of optim l movement of the objective function
d) If two or more constr ints re p r llel to e ch other 28. Three persons enter
into r ilw y c rri ge nd there re 8 se ts v il ble. In how m ny w ys they
c n se t themselves? ) 24
b) 336
c) 40 d) 56

29.
In which of the following the simple h rmonics me n is ppropri te: ) A set of
r tios using the numer tors of the r tio d t s weights b) A set of r tios usin
g the denomin tors of the r tio d t s weights
c) A set of r tios which h ve been c lcul ted with the s me numer tors
d) A set of r tios which h ve been c lcul ted with the s me denomin tors
30. Which of the following st tements is not true bout st nd rd devi tion? ) C
ombined st nd rd devi tion of two or more groups c n be c lcul ted
b) The sum of the squ res of the devi tions of items of ny series from
v lue
other th n the rithmetic me n would lw ys be sm ller
c) St nd rd devi tion is independent of ny ch nge of origin d) St nd rd devi ti
on is dependent on the ch nge of sc le 31. Which of the following is/ re true wi
th respect to geometric me n :( ) Geometric me n c nnot be c lcul ted if ny of
the v lue in the set is zero. (b) Geometric me n is ppropri te for ver ging th
e r tios of ch nge, for ver ge of proportions, etc. (c) Geometric me n is consi
dered most suit ble ver ge for index numbers.
Only (I) bove
(i) Only (II) bove (ii) All (I), (II) nd (III) bove (iii) Only (II) bove 32.
The prob bility of getting two he ds from three tosses of f ir coin is- ) 1/8
b) 1/4
c) 3/8
d) 1/2 33. If A nd B re two mutu lly exclusive events
prob bility of events A nd B h ppening together is--

nd P(A) = 2/3, then the

) 0
b) 1/3 c) 2/3 d) 1/2 34. Which of the following c n be directly used s the test
st tistic in hypothesis tests on the b sis of non st nd rdized sc le :( ) The s
mple me n, when the test involves the popul tion me n. (b) The difference betwe
en two s mple me ns, when the tests involve the difference between two popul tio
n me ns. (c) The s mple proportion when the test is bout the popul tion proport
ion (i) Only ( ) bove (ii) Only (b) bove (iii) Only (c) bove
(iv) All ( ), (b), (c) bove
35. A box cont ins 60 b ll point pens out of which 10 pens re defective. 8 pens
re r ndomly picked up from the box. The prob bility distribution of the number
of pens which re r ndomly picked, will be- ) A discrete uniform distribution

b) A binomi l distribution
c) A hyper geometric distribution d) A Chi- squ re distribution 36.
er the process of selecting simple r ndom s mple s n experiment
f the following c n be tre ted s r ndom v ri ble(s)? ( ) S mple me
st nd rd devi tion (c) S mple r nge (d) S mple medi n (i) Only ( )
nly (b) bove

If we consid
then which o
n (b) S mple
bove (ii) O

(iii) All ( ), (b), (c), (d) bove

(iv) Only (d) bove 37. The cov ri nce of r ndom v ri ble with itself is lw ys) A positive qu ntity
b) A neg tive qu ntity c) 0 d) Less th n its expected v lue 38. A m n h s 6 frie
nds. In how m ny w ys he c n invite one or more of them to p rty?
) 63
b) 64 c) 119 d) 120 39. Find x; if logx/log2 = log36/log4 ) 0 b) 2 c) 4
d) 6
40. The empiric l rel tionship between r nge (R) nd me n devi tion (M.D) is- )
2R=15M.D
b) 3R=17M.D
c) R=17M.D d) 3R=M.D