” oO - vet Uoyr Example 4. The following table gives the age of cars of a certain make and annual maintenance costs. Obtain the regression equation for costs related to age : Age of Cars (in years) 5 2 4 6 8 Maintenance cost (in @ hundred) : 10 20 25 30 Also estimate the annual maintenance cost for a ten year old car. Solution. Let X denote the age of car and Y denote its annual maintenance cost.Example 5. A furniture retailer in a certain locality is interested in studying the relationship that exists between the number of building permits issued in that locality in the past years and the volume of sales in those years. The data for sales (¥, in thousand rupees) and the number of building permits issued (X, in hundreds) for the last 10 years are collected to give the following results : EX= 200, Y¥=2200, YXY=45800, yx? =4600, yy?= 490400 Find : (9 the level of sales expected in a year when 2000 building permits are to be issued, and (i) the change in sales expected for every increase of 100 building permits.Example 6. Given X =50, ¥ (i, Correlation coefficient, Fag. Jer NU VOrrelauon = 20,6, = 20, o)= 20 and Cov (X, Y) = -100. Find: (i) both the regression coefficients \Dethi Univ. B.Com. (P) 2014 (SOL me las nes NA. ands incon DR9.12 Example 7. Find the regression. coefficients byx respectively, if standard deviations of Xand Yare 4 of correlation between X and Y is 0.8. Bat WIA nnn wien the fallaurine + Business Statistics and by, of Yon X and X on Y and 3 respectively, and coefficient= UOX> — UU am KY 3 Py 4 os Example 8. The coefficient of correlation between ages of husbands and wives in a community was found to be +0.8, the average of the husbands’ age was 25 year: Sand that of wives’ age was 22 years. Their standard deviations were 4 and 5 respectively, Find with the help of regression equations : () The expected age of the husband when wife’s age is 18 years, and (i The expected age of the wife when husband’s age is 32 years. [Dethi Univ. B.Com. (H) 2010)X= 521M (1). LNUS, EXpeurcu age vi wae we Example 9. Find most likely production corresponding to the rainfall of 40 cms from the following data using a regression equation : Rainfall Production Average 30 cm 500 kg Standard deviation 5cm 100 kg Coefficient of correlation 0.8 [Dethi Univ. B.Com. (P) 2009) Solution. Let X denote the rainfall fin em) and V denote the nraduction (in kg). Thepunury example 10. The following data are given regardin i isi ates pesticular firm gi g expenditure on advertising and "© -Y Mee se UE UU Bo UUU Dg: Adv. Expenditure (X) Sales (Y) (lakhs) (#lakhs) Mean 10 90 Standard deviation 3 12 Correlation coefficient 08 (j Calculate the regression equation of Yon X. (i) Estimate the advertisement expenditure required to attain a sales target of 120 lakhs. [C.A. Foundation, Nov. 1999] awe wr ateeee tha fallaurine information :A= Litera regression soueene e value of X when Yis 130 and estimate the value of Y when X is 30. x : 39 33 30 31 32 36 41 49 46 43 Y : 132 134 138 129 136 131 132 135 128 125 [Dethi Univ. B.Com. (P) 20. 12] Solution. Calculations For Regression Lines‘ seanple 15. The regression coefficient ic g : of regression equation of X i the same for regression equation of Y on X is 0.8. Are the een Ys eae [Dethi Univ. B.Com. (F) 2012] consistent?FAS ULE Lae Lg EOIN: Cea ES te Example 17. Given below is the information relating to a bivariate distribution: \kegression equation of Yon X. Y= 20 + 0.4X, Mean of X = 30, Correlation Coefficient between X and Y = 0.8. Find regression equation of X on Y. (Dethi Univ. B.Com. (H) 2005 (SOL), 2019,> An ee _peimote 20. A departmental store gives in-service training to its salesman wh followed by a test. The management is considering whether it should terminate” is service of any salesman who does not do well in the test. The following da ta giv © the test scores and sales by nine salesmen during a certain period : €S the Test scores 14 19 24 21 26 22 15 20 Sales (‘00 €) : 31 36 48 37 50 45 33 41 Calculate the coefficient of correlation between the test scores and the sales. indicate that the termination of services of low test scores is justified? If 4 wants a minimum sales volume of 3000, what is the minimum test score that wil} ensure continuation of service? Also estimate the most probable sales volume of salesman making a score of 28. [Delhi Univ. B.Com. (H) 1998, 2003, 20; 4 Onnes new lae to Sia 19 39 Does it he firm |. wee wenens AUG CULE UE Ua wee Exdeple 21. Find the two regression coefficients for the data given below : mo: 3 5 5 4 8 12 Nee 18 14 12 10 9 7 Let now each value of X be multiplied by 2 and then 6 be added to it. Similarly, multiply each value of Yby 3 and subtract 16 from it. What will be the new regression coefficients between the new series of Xand Y? Solution. Computation of Regression CoefficientsA nay mee ee eee Settee wate Bkample 23. A panel of Judges A and B gre ‘aded seven debators and independently awarded the following marks : bey Debator a er ee fra ae, MarksbyA : 40 34 28 30 44 38 31 Dx Ov, Marks by B > 32 39 26 30 38 34 28 An eighth debator was awarded 36 marks by Judge A while Judge B was not present. If Judge Bwere also present, how many marks would you expect him to award to the eighth debator assuming that the same degree of relationship exists in their judgement? [Delhi Univ. B.Com. (H) 1993]you are given the following information : _ pane 24. For a bivariate data, ¥(xX-58) = 46, L(¥-58) =9 ¥(X-58)? = 3086, 483, > (X-58)(¥-58) = 1095 Wl y(y-58)? No. of pairs of observations = 7 You are required to determine (i the two-regression equations and (i) the coefficient of correlation between X and Y series. [Delhi Univ. B.Com. (H) 2000, 2017, 202]sf 25. You are given the following information regarding a distribution : N=5, X=10, ¥ =20, X(X-4)? = 100, ¥(¥ -10)? = 160, E(X-4)(Y-10) = 80. find the two regression coefficients and hence the coefficient of correlation. [Delhi Univ. B.Com. (H) 2002] solution. If we let U= X-4 and V= Y- 10, then we are given : yu? = 100, LV? = 160, Luv =80 Also, DU = O(X-4) = DX-4n = wX-4nw = 5x 10-4%5 = 30 and XV = L(¥-10) = YY-10w = n¥-10N =5« 20-105 = 50 Since the regression coefficients are independent of change of origin, therefore, b= by = MEW -(ZU)(EV) _ 5x80-30x50 wow nyu? -(Su)> 5 x100-(80)? _ 400-1500 _-1100 _ 11 500-900 ~ -400 ~ 4 anil byy' Boy NYUV -(LU)(ZV) _ 5x80-30x50 NYV?-(Sv)? ~ 5x160-(50)" ~ 400-1500 _ -1100 11 ~ 800-2500 ~ -1700 © 17 Coefficient of correlation (1) is given by r= Ybxyxbyx = ftextt = 2 = 1334. But rcan't exceed 1. Hence the given data are inconsistent. 1 Bim eee = rerro 1 yras - ‘obtained : R-5, ¥ =4, yX? = 680, yy? = 480, XY =508 It was later on discovered that two pairs of observations (X, Y) were noted as (8, ii and (5, 8) while correct values were (10, 12) and (3, 6) respectively. Determine tt: corrected values and find :ea assion Analysis ~ i i 9.21 a the two regression coefficients, a the correlation coefficient, and when supply = 80 units and the supply when price = 70. ( 1996) gga the price [Dethi Univ. B.Com. a yx < = v -n¥Business ati, 9.28 class, the regression e 7 Z, sple 28, For 100 students of a clase, ce 4 00 tte in Economies (¥) is 3Y~ 9X + 180 = 0, BOP of may fmarks in Statistics is 4/9 of gan yeh the vars, ath, Statistics (X) on the r n in Economics is 50 and variance 0} Cee marks in Economics. Find the mean marks in Statistics and the cont correlation between marks in the two subjects. [Dethi Univ. B.Com, cient fe et : . 9 regression line of Xon Yis given by : 3Y-5X+ 189 ea 20031 Solution. The . 5X = 3Y +180 [ Since the lines of regression always pass through the point (X, 7), “ty X= BY +180=3% 50 + 180= 330 i it F wig = xz - 322 = 66 % Thus the mean marks in Statistics is 66. 2 402 o% _ 4 cx 2 i : Rag 727 =9 —~ == Further, we are given: ox 9 oy or of 97 sy 7 - 3 Solving Eq. (1) for X in terms of Y, we obtain: X = ZY +36 = Regression coefficient of Xon Yis: byy = 2 3 253 _9o | r= 7p 709 ; rok o ie, a i Thus the coefficient of correlation between X and Yis 0.9. Rit tata thn Han af ronrescion of ¥ on X and that of X on ¥ out of te_ pene 30. Out of the following two regression lines, find the line of regression of 2 on Y: QX+3Y=7 and SxX+4Y=9 oe o: ieeeNS? MKT 2 ~ Gx 2 7 2B DT le 33. If the two regression lines are 3Y + 9X= 46 and 3X+ 12Y=19, determine “hich one of these is the regression line of Y on X and which one is that of X on Y. Also find the means, correlation coefficient and ratio of the variance of X and Y. [Dethi Univ. B.Com. (H) 2006 (SOL)vy ' iven X= 4Y+5 and Y= kX+ 4 are the lines of regression of Xn y, 1 Sng phage 34.6) Yon X respectively. If kis positive, prove that it cannot exceed 3. Ifk= 75, tnd, means of two variables and the coefficient of correlation between them, [Dethi Univ. B.Com. (2) 2096 Solution. The regression line of X on Y is given by X=4Y¥+5 => dy=4 The regression line of ¥ on Xis given by Y-=kX+4 > by=k 12 = byy * by = 4k - r2< (always) = 4k<1 => % ke $< Ifk = Z&, then the regression line of Yon Xis : : 1 1 Y= GXt4 > by 76 = by bye 4xce ed Spe at xy: Oe 16" 4 crak 2? the positive sign is being taken, because r, byy and by, all have the same signs. Since both the lines of regression pass through the mean values (X,Y), therefore Site and ¥ =7¢X+4 wa (2) X =47+5 . (Q) Solving Eqs. (1) and (2) simultaneously for X and ¥, we obtain X = 28 and ¥ =5.75./ 9.35 _ sian 95. Find the us of regression of Yon X from the following data : ie : 8 10 12 14 uf : _ . 5 10 4 uw also calculate the standard error of estimate of Yon xX, solution. TABLE 9.1. Calculation for Regression Line x x ue aa XY Yo | Y-¥, WV 2 2 ay 16 8 Fane 1.1784 1.3886 A 3 16 9 12 4.2144 -1.2144 1.4748 6 5 % 25 30 5.6072 -0,6072 0.3687 8 10 64 | 100 80 7.0000 3.0000 9.0000 10 4 | 100 16 40 8.3928 ~4,3928 19.2967 2 us| 14a | aan 132 9.7856 1.2144 1.4748 4 12 | 196 | 144 168 11.1784 0.8216 0.6750 rx | LY | ox? | yy? | yxy Z¥e | L¥-¥c) | Xer-¥e)? =s6 | =49 | =se0] =431 | =470 =49 =0 = 33.6786 Note that )}Y and }\Y¢ are equal. This must always be true if L(¥-Ye) =0. 7 Y 4 x. 2X. and y- 2.9 ., AY XY -(EX)\(EY) _ 7x470-56x49 ~ ex? -(Ex)? 7560 -(56)” ss 3290-2744 _ 546 = = 3920-3136 ~ 784 ~ 06964 Regression equation of Y on X is given by y-Y= byx (X-X) or, Y-7= 0.6964 (X-8) > Y = 1.4288 + 0.6964X = (a) To calculate the standard error of estimate (of Yon X), we must compute a value for Y, for each value of X by substituting each X value into the regression equation (1). We must then compute the difference between these Y, values and the corresponding values of Y. These values are shown in Table 9.1. ‘The standard error of estimate of Y on X is calculated as follows : 2 E(¥-Yo)” _ [836786 _ ais 21934 Sy 7 7 Zaress analysss 9.37 Beg 98. Given the following data : (mp 6 4 7 4 8 9 6 9 § 9 rate: (9 the regression coeff Coane ea Sennen and (iv) the s' ‘ 3 1006 8 7 9 2 6 8 7 UW 8 7 It cient of Y on X, (ii the total variation in Y, (ii) the landard error of estimate of Yon X. Computation of Regression Coofficiont ton eels ye xy Yo ¥-¥, |v- exe vet IF 4.29 + 0.53x| = t : roy-1 TAT 153 | 2.3409 38 Y = 8.805 + 1. 15X When price is 16 units, thé Sstimatéd supply is obtained by substituting X= 163 4 Vig = 8.805 + 1.015 (16) = 25.045 Sint ) is given by Y— 22 = 1.015 (X- 13) The standard error of estimate (of Yon : [5506 ~ 8.805 (220) -1.015(3467) ieee a = 8.805 and b= 1.015) Fi 0 _ _/5506-1937.1-3519 _ [49.9 06 199718518 A > Va99 = 2.234. 9 salesmen of @ 7 : Beam mple 40. The following data relate to the scores obtained by ‘ompany in an intellignece test and their weekly sales performance: B oc bp & F GH ! S, yalesmen Sh ‘est Scores 7 :@ 50). 60, #50 60 80 in' 50 80 40 2% eter ica 30 60 40 50 60 30 70 30 © ’) Obtain i : the regression equation of sales on intelligence test scores of the salesme™ ted week ii) If the intelligence se ‘ sales? ore of a salesman is 65, what would be his expec ) Cal i) Calculate the standard error of sales on intelligence test scores. od [Dethi Univ. B.Com wy? adle 42. Calculate the standard error of the estimate of Y from the regression of on X for the following data : 2 C n=10, Ly =90, Lxy=120, yx? =200, where x= X-X, y= ¥-¥ 7 (Det Univ. B.A. (Econ: Hons.) 1991] 2 a ee a ees f Van Vie\_Bkample 44. The equations of two lines of regression, obtained a 4 Correlation 2... between variables X and Y (not necessarily in that order) are as follows: P aalisy 2X*+3Y-8=0 and 2Y+X-5=0 ‘The variance of X = 4. Find : {Q Nariance of Y, (ii) Coefficient of determination of Xand Y, and (iti) standard em, oi estimate of Y on X. [Delhi Univ. B. Com. (H) 2004, 2011, 2012, 2022(Modified Satutian The ennatione of twa resression lines are :