0% found this document useful (0 votes)
35 views3 pages

Statistical Problem Set on Central Tendency

The document contains a problem set with 20 questions related to statistics and mathematical inequalities. Topics include the behavior of means and variances under certain conditions, properties of central tendency measures, and applications of inequalities like AM-GM and Cauchy-Schwarz. It also explores the implications of transformations on data distributions and the relationships between different statistical measures.

Uploaded by

techiesaditya
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
35 views3 pages

Statistical Problem Set on Central Tendency

The document contains a problem set with 20 questions related to statistics and mathematical inequalities. Topics include the behavior of means and variances under certain conditions, properties of central tendency measures, and applications of inequalities like AM-GM and Cauchy-Schwarz. It also explores the implications of transformations on data distributions and the relationships between different statistical measures.

Uploaded by

techiesaditya
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Problem Set 1

1. Suppose there are two groups of observations with means M1 and M2 , respectively. When
some observations from group 1 are transferred to group 2, the new means for these two
groups turn out to be M1∗ and M2∗ , respectively. Discuss whether it is possible that the two
inequalities M1∗ > M1 and M2∗ > M2 hold simultaneously.

2. Suppose there are n observations with mean x̄ and variance σ 2 . Can you add one new
observation to this dataset such that the mean and the variance both remain unchanged? Can
you do that if you are asked to add two new observations?

3. Two groups of observations have means M1 and M2 , respectively. If M denotes the mean
of the combined dataset, show that M lies between M1 and M2 . Does this result hold for
median and mode as well? Justify your answers.

4. Let m1 , m2 , m3 be the mean, median and mode of the daily maximum temperatures in
centigrade scale computed over a period of 30 days in Kolkata. If the observations are
measured in the ferhenhite scale, what will be the mean, median and mode?

5. Let m1 , m2 and m3 be the mean, median and mode of a frequency distribution with mass
points x1 , x2 , . . . , xn and the corresponding frequencies w1 , w2 , . . . , wn . Define yi = f (xi ) for
i = 1, 2, . . . , n, where f is a monotone but nonlinear function. Let M1 , M2 and M3 be the
mean, median and mode of the frequency distribution of the transformed data. Check whether
the relations M1 = f (m1 ), M2 = f (m2 ) and M3 = f (m3 ) always hold.

6. The mean and the variance of the scores of 10 students were recorded as 60 and 16, respectively.
Later, it was found that one observation, which was originally 70, was wrongly recorded as 50.
Compute the corrected mean and variance of these 10 observations.

7. Use appropriate measures of central tendency to answer the following questions.

(a) A person travels the first 40 km at a speed of 40 km/h, the next 60 km at a speed of 60
km/h, and comes back the entire distance of 100 km at a speed of 100 km/h. What is
the average speed over the entire journey?
(b) The price of a company share grows by 10% in the first year, 20% in the second year,
and 30% in the third year. What is the average percentage growth rate over these three
years?

8. Can you comment on the breakdown point of (i) mean, (ii) median, (iii) standard deviation,
(iv) mean deviation and (v) quartile deviation?

9. Discuss how you will find the mode of a grouped frequency distribution from an appropriate
histogram when the classes are not of the same width. Can you find the median from this
histogram? Justify your answer.

1
10. Consider to sets of observations {x1 , x2 , . . . , xn } and {y1 , y2 , . . . , yn }. Discuss how you will
find the minimizer of the function
n
X
ψ(θ) = |yi − θxi |.
i=1
Pn
[You know how to find the minimizer of i= wi |xi − θ|. Try to use that knowledge.]

11. Using the AM–GM inequality, show that if the perimeter of a rectangular field is 36 meters,
its area cannot be bigger than 81 square meters.

12. If x1 , x2 and x3 are three positive real numbers such that x1 + x2 + x3 = 1, show that
8
8x1 x2 x3 ≤ (1 − x1 )(1 − x2 )(1 − x3 ) ≤ .
27

13. Let a1 , a2 , . . . , an be n distinct positive real numbers and b1 , b2 , . . . , bn be a permutation of


a1 , a2 , . . . , an . Prove that
a1 a2 an
+ + ··· +
b1 b2 bn
attains its maximum value when ai = bi for all i = 1, 2, . . . , n.

14. If x, y and z are three positive real numbers.

(a) If x + y + z = 1, find the maximum value of x2 y 3 z 4 .


(b) If 2x + 3y + 4z = 1, find the maximum possible value of xyz.

15. Show that the mean absolute deviation about the median cannot be larger than the standard
deviation. Hence or otherwise, show that the difference between the mean and the median
cannot exceed the standard deviation.

16. If n observations lie in the interval [0, t], show that variance cannot exceed x̄(t − x̄). When
does the equality hold?

17. If there are n observations x1 , x2 , . . . , xn , the range is defined as

R = max xi − min xi .
1≤i≤n 1≤i≤n

If s2 denotes the variance of these n observations, prove that R2 /2n ≤ s2 ≤ R2 /4.

18. (a) If you are asked to place 10 observations x1 , x2 , . . . , x10 on the interval [0, 1] so that the
variance is maximized, how will you do that?
(b) Given that you have already placed the 10 observations in optimal places (following (a)),
if you are asked to place a new observation x11 , what will be your choice for x11 ? Justify
your answer.

19. (a) Consider two sets of observations {x1 , x2 , . . . , xn } and {y1 , y2 , . . . , yn }. Define Sx =
{ ni=1 x2i }1/2 and Sy = { ni=1 yi2 }1/2 . Now, using the inequalities
P P

n
!2 n
!2
X xi yi X xi yi
− ≥ 0 and + ≥ 0,
i
Sx Sy i
Sx Sy

2
Pn
prove that | i=1 xi yi | ≤ Sx Sy , and hence

n
!2 n
! n
!
X X X
x i yi ≤ x2i yi2 .
i=1 i=1 i=1

This is known as the Cauchy-Schwarz (CS) inequality. When does the equality hold?
(b) Using this CS inequality, for any n real numbers a1 , a2 , . . . , an , prove that ni=1 a2i ≥
P

( ni=1 ai )2 . (You can also prove this result using the simple fact that ni=1 (ai − ā)2 ≥ 0,
P P

where ā is the arithmetic mean of a1 , a2 , . . . , an ).

20. Let A, G and H be the arithmatic mean, geometric mean and harmonic mean, respectively, of
n positive numbers x1 , x2 , . . . , xn . Using the fact that ex−1 ≥ x for all x, or otherwise, prove
that A ≥ G ≥ H. When does the equality hold?

You might also like