Business Soha S. C. GUPTA INDRA GUPTA Himalaya Publishing House© No part of this publication should be reprodiiced, stored in a retrieval system, of transmitted in a form or by any means, electronic, mechanical, photocopying, recording and/or otherwise wenke yn? prior written permission of the author, Breach of this will be liable for legal action } aaa First Edition June 1988 Fifth Thoroughly f Revised Edition: January 2005 Reprint : 2007 —_—_ Published by Mrs. Meena Pandey for HIMALAYA PUBLISHING HOUSE, “Ramdoot”, Dr. Bhalerao Marg, Girgaon, ‘Mumbai-400 004. Phones : 23860170/23863863 Fax : 022-23877178 Email : [email protected] Website : www.himpub.com Branch Offices Delhi : “Pooja Apartments’, 4-B, Murari Lal Street, Ansari Road, Darya Ganj, New Delhi-110 002 Phones : 23270392, 23278631 Reliance 3018092 to 96 Fax : 011-23256286 5 Email : [email protected] Negpur : Kundanlal Chandak Industrial Estate, j Ghat Road, Nagpur-440 018 Phone : 2721216, Telefax : 0712-2721215 Bangalore + No. 16/1 (old 12/1), Ist floor, Next to Hotel Highland, Madhava Nagar, { Race Course Road, Bangalore-560 001 Phones : 2261641, 2886461 Fax : 080-2286611 Hyderabad + No, 22-1 167/2H, 1st Floor, Near Railway Bridge, Tilak Nagar, ‘Main Road, Hyderabad-600 044 Phone ; 26601745, Fax : 040-27560041 ‘Typeset at + Blite-Art, Pataudi House, Daryaganj, New Delhi-110002 Printed at + New Offset Peinters New Delhi 2 pene oe OM NEE @ scanned with OKEN ScannerAVERAGES 5.47 5,9. GEOMETRIC MEAN The geometric mean, usually abbreviated as G.M.) of 3 product. Thus if Xj, Xo... Xq are the given m obs set of n observations is the n'® root of their vations then their G.M. is given by GM VX XK K Ky = OK XD XQ (521) Ifn=2 ie,, if we are dealing with two observations only then G.M. can be computed by taking the quare root of their product, For example, G.M. of 4 and 16 is Vax t6= Vor = 8 re ve ane of observations is greater than 2, then the computation of the n'™ root is very tedious. In such a case the calculations are facilitated by making use of the logarithms. Taking logarithm of poth sides in (5:21), we get log (G.M.) = Hog (X)X2...X,) =} (log X; + log Xp +... + log X,) ne logX (SD) ‘Thus we see thar the logarithm of the G.M. of a set observations is the arithmetic mean of their logarithms Taking Antilog of both sides in (5-21a), we finally obtain, GM. = Antilog| ?¥ log xX] (5-216) In case of frequency distribution (X;,f); @= 1, 2, ..., n, where the total number of observations is N=3f, GM. 5 [41% Xp sf times) x (Xp x Xa X fates) x «x XX XX «fatimes |!” = (XSi x Xpf2x x X_N e-(5:22) ‘Taking logarithm of both sides in (5-22), we get log GM. =4[ log (X41. X2.. x,fa| [log Xfi + log Xo + weblog X,fn] es =h = [flog X +frlog Xo+ s+ falog Xn] = PE floeX 1: (5-224) = GM. = Antilog [7 /loz x] ve (5-226) In the case of grouped or continuous frequency distributions, the values of X are the mid-values of the corresponding classes. Steps for the Computation of G.M. in (5-225) 1. Find log X, where X is the value of the variable or the mid-value of the class (in case of grouped or continuous frequency distribution). 2. Compute fx log X i.e., multiply the values of log frequencies. 3. Obtain the sum of the products flog X obtained in step 2 to get Lflog X. 4. Divide the sum obtained in step 3 by N, the total frequency. 5, Take the Antilog of the value obtained in step 4. The resulting figure gives the value of G.M. X obtained in step 1 by the corresponding 59-1, Merits and Demerits of Geometric Mean, Merits : (i) Geometric mean is rigidly defined, @ scanned with OKEN Scanneruns ae 7 5 iste na he ober jal wreatment. IF Gand Go are the ea wi 0 for further mathemalt rn the geometric mean G of the combin Meany Ha) Tan dn roapectivel) ce (eB on groups of sires m ty SE a mn lop Git M2 108 Ga loge ah «syean be easity generated 10 the 6086 Of k BEOUPS as Follow ) esult in (5:23) eee Remark, The resi os geometric means ofthe K groups of SiS. 2... my mac aren sorte mbined group of size ny +2 +... + ‘ soomstric mean G of the combi ‘ my og Gy + na log Gy + +108 Ge ae aaa “2 ‘ for higher values, geometric mean has biag Unlike arithmetic mean which has @ bias , a dooaneeeaae ic gute useful in phenomenon (such as prices) which has a lowes ated oi go hslow zero) but has no such upper limit. : oo none wih mean, G.M. i affected toa lesser extent by extreme observations, (oi Tes not affected much by fluctuations of sampling. : 5 Demers.) Because ofits abstract mathematical character, geometric mean i not easy 0 nde and to calculate for a non-mathematical person. 1) A any one of the observations is zero, geometric mean becomes zero and if any one of Sbservatons ss negative, geometric mean becomes imaginary regardless of the magnitude of the aie stems ‘ses inspite of its merits and limitations, geometric mean is specially useful in averaging rao, percentages, and vates of increase between two periods. For example, G.M. is the appropriate average eke ssc’ for computing the average rate of growth of population or average increase in the rate of profits, sec Production, etc., or the rate of money. comeiric mean is used in the construction of Index Numbers. Irving Fisher's ideal index numbers ‘based on geometric mean [See Chapter 10 on Index Numbers). While dealing with data pertaining to economic and social sciences, we usually come across the Stwations where it is desired to give more weightage to smaller items and small weightage to larger items. Mts the most appropriate average to be used in such cases, $2. Compound Interest Formula, Let us suppose that ois the initial value of the variable (ie. te alue of the variable inthe beginning and P, bets value at the end of the period n and let r be the rate of growth per unit per period. 5 Since ris the rate of growth per unit per period, arate atthe end of period 1 is r Py Py = Py (1 becomes FoI +). The growth forthe 2nd peiod fe at the end of 2nd period is growth for period 1 is Pp r and thus the value as 1). For the 2nd Period the initial value of the var Po(1 +r) rand consequently the value of the varia! Py = Po(l +P tr Similarly proceeding, the value ofthe vai Pr= Pol +r} + Po(1 +p), and finally, its value at the end of petiod n w P,=Po(l try, which is the compound interest formula Ptr (ltr Pol tre ble atthe end of period 3 is T= Po(1 +r)[1 + 7] =Po(1+r) ill be given eH) for money, ves four unknown quantities Fy The value atthe end of period a: The length of she peri "o"The Value in the beginning + 1 The rate per unit Per period, y Equation (5-24) iny @ scanned with OKEN ScannerEe vena 5.49 spwo ae given Po? and n we can compute 2, by using (5:24) directly. However, (5-24) can be used (0 rwinany one cof the four values when the remaining three values are given, Ns nk Average Rate of Variable Which Increases by Different Rates at Different Pertods. Lt us anos ited of the values of the varlabe increasing ata constant rate in each period, the rate Pot period is different, 8AYs 7, Posty FOF the Ist, 2nd,...and nth ‘Then, as inthe previous seation We shall get See A aioge a vit Pet goss P, = The value at the end of Ist period = Py (1 +r) Py = The value at the end of 2nd period = Pp (1 +7) (1+ 79) p, = "The value at the end of period = Pdr) (1+ 1) + hy) oe) trys assumed to be the constant rate of growth per unit per period, then we get P,=Po(l+n" Atay, Hence, equating the values of P, in (*) and (*), the average rate of growth over the period n is given y dant =n) C4). 4m) 1m 4 ler =[C4n) 047)... +n)] I | ry denote the percentage growth per unit per period for the n periods respectively then we rede [lt] (Onl 20 here ris the average percentage growth rate over n periods. have 100+r = [(100 +r) (100+ ra) (400+ %)) [ca00 + 71) (100 + 72) «(100+ r,)]." = 100 .(5-26a) ‘Thus we see that if rates are given as percentages, then the average percentage growth rate can be sbiained on subtracting 100 from the G.M. of (100 + r1),(100 + rp)--(100 + tm). Remark. It should be clearly understood that average percentage growth rate is given by (5:26) and not ty the geometric mean Of 1, Fa...» Tar > r 54:4, Wrong Observations and Geometric Mean. Let us suppose, that the value of the geometric mnean computed from n observations, say, Xj, X2s--0Xn is G- ‘On checking, it is found that some of the observations, say, X;, Xp and X, were wrongly copied instead of the correct observations X;’, Xo" and Xs Weare interested in computing the correct value of the geometric mean, G = Geometric Mean of Xj, Xav +o Kn= (Xi Xo - Xa-oXy)it -C) Xp and X; by the correct values, the corrected value of the On replacing the wrong observations Xy, ‘feometric mean, say, G’ is given by Ue C2 (hy Ky Bylo X= ( ort Xt Xot Xylyuln = (y Xy Kars Be) Hea a (Ee) Grom Ch Guaet 1X Xs The result in (5-26), can be generalised to the case of more than three observations. For illustration, *e Example 5:34, = @ scanned with OKEN Scanner. 7 “ Geometric Mean’ of 24,8, 12, 16 and 2 % rind the Gi Bynmple $92 Fi Solution, X ch 5 134805 log N 469 (Using (5.21 oy as ou16 8158 0.9116) = #15 GM. = Antilog ( Teer GM, = (2x4xBX 12x 16x 24)! = ( ws SM, =hiog 294912 = 20% = 09116 log GM. = blog 294912= 5 , = 8158. GM. = Antilog (09116) ve crtbation: a ic the following distribution : 38, Find the geometric mean for anne 0-10 10-20 20-30 30—40 oy weet 7 15 25 ; No.of students 5 Solution. Marks Mid-Point(X) No. of Students () log X Tox 10 Ss 5 06990 eer hon _ 7 1.1761 ae 20-30 2 15 13979 aaa 38 = 15441 aaa a" : 8 1.6532 ie “se 845083 Geometric mean = Antilog [ 2px] = Antilog | ssa | = Antilog [1-40874] = 25:64 marks. Example 56. The geometric mean of 10 observations on a certain variable was calculated as 162.4 was later discovered that one of the observatic ions was wrongly recorded as 12-9; in fact it was 21-9. Ap appropriate correction and calculate the cor rrect geometric mean. Solution. Geometric mean G of n observations is given by : GX Kaye Ket > Gm =X Xp .Xq au Thus the product of the numbers is given by Xk... X= G*= (16.2910 (Given n= 10, G = 16.2) | tthe wrong observation 1259 is replaced by the correct, om ‘alues 21-9, then the corrected value of product of 10 numbers is ob vidi jon in fl i i Pret of 1 numbers a Thor ©" dividing the expression in *) by wrong observation and mulls Corrected product (XiX...X,) =Msaxaig Hence, comrected value ofthe geometric mean G’, (say), is given by : o [scapeeatay™ : * Ta el log G' = [108 16:2) + 19219 top 129]= 1 1 T5[ 10 log 162 + 1og21-9 - tog 129] =ih[ 10x 1.2095 4 13404-11106] 1 10 12.0950 + 1.3404 = 12325 106] tol 134354 1-196) = 2a © = Antilog (1-2325) = 17.08 @ scanned with OKEN Scanner“ AVERAGES 557 ren, Average speed = Total distance travelled _ 81+ 6+... +5 Total time taken 144... 44 HASH ts re i) Xs 1% 1 [From (*)} ...(5-30) oy 5 I 5 ‘al at) (3)2(>) which is the weighted harmonic mean of the speeds, the corresponding weights being the distances covered. Hence, Waiforentdinences are traveled with corresponding different speeds, then the average speed js given by the weighted harmonic mean of the speeds, the corresponding weights beings the distances Byamaple $47, You make a trip which entails travelling 900 kms. by train at an average speed of 0km. pfs 3000 kms. by: boat at an average of 25 km. p.h.: 400 kms. by plane at 350 km. p.h., and finally S km. p.h. What is your average speed for the entire distance? ys kms by taxi at 2 MPUTATION OF WEIGHTED HM. Solution. Since different distances are covered with! x Ww Wik varying speeds, the required average speed is given by the a 1 Veiehted harmonic mean of the speeds (in km. p.h) 60,25, 350] °° ne 8 nd 25: the corresponding weights being the distances covered| 25 || | °0 1 din kms.) vic, 900, 3000, 400 and 15 respectively. oa ae mn Average Speed = p Siti = 315 = 31-49 km. ph, W=4315 | EOWA) = 137-03 | sit. RELATION BETWEEN ARITHMETIC MEAN, GEOMETRIC MEAN AND HARMONIC The arithmetic mean (A.M.), the geometric mean (G.M.) and the harmonic mean (H.M.) of a series of n observations are connected by the relation : AM. 2 GM. > HM. (5-31) the sign of equality holding if and only if all the n observations are equal. Remark. For two numbers we also have @=AxH (5-34) where A, G and H represent arithmetic mean, geometric mean and harmonic mean respectively. Example 5-48, H.M., A.M. and G.M. of a set of 5 observations are 10-2, 16 and 14 respectively.” ‘Comment. . Solution. We are given : n = 5; AM. = 16; GM. = 14 and HM, = 102. Since AM. > GM. >HM., the above statement is correct. Example 5-49. The arithmetic mean of two observations is 127-5 and their geometric mean is 60. Find (i) their harmonic mean and. (ii) the two observations. Solution. (i) Let the two, observations be a and b. Then we are Arithmetic Mean =2*%= 1975 = a+b=255 To) GM. =Vaxb=60 => ab = 3600 wo) Harmonic mean of two numbers a and b is given by : 2ab__ 2x 3600 _ 480 _ 4.54 [From (*) and (**)] EY op ess Aliter. For two numbers, we have : @ =AH = (ii) We have (a= by? = (a+ b)*—4ab = (255)? — 4 x 3600 [From (*) and (**)) 65025 ~ 14400 = 50625 @ scanned with OKEN Scanner