rrr Correlation Analysis: KARL PEARSON'S COEFFICIENT OF CORRELATION FORMULAE Karl Pearson's Co-efficient of correlation 1. When deviations are taken from actual mean _ By Ldud, © Ngxsy Nixsy ty Zdxdy r or) ee Vaex2 Sax? - Vea? 2. When deviations are taken from assumed mean eae Zdx) (Edy) Bos ty n= Dds? — (Sd)? Vn- Day? — (ay)? 3. Probable Error: PE. =0.6745 + g VN 4. Standard Error : = Karl Pearson's Coefficient of Correlation =(X-X) (Y-Y) N = Number of Pairs of observations $, = Standard Devation of X series Sy = Standard Devation of Y series x. pS x = Actual mean of x series y = Actual mean of y series. = (X-A) =i (Yeuay A. = Assumed Mean of X and Y series P.E. = Probable Error r = Correlation N_ = Number of pairs of observation eee = Standard Error SOLVED PROBLEMS TYPE NO.1 Problem : 1 Find out thé*co-efficient of co-relation from the following data. Price (in) SAPONINS Moot 75) [oss2 "190 Lent eees | 39 Supply (inKgs): | 85 | 51 | 91 | 60 | 68 | 62 | 86 | 58 | 53 | 47 Solution : Calculation of co-efficient of co-relation 5 de ae y dy a dxdy | 78 16 256 85 25 625 400 36 -26 616 SI 3 81 234 98 36 1296 91 31 961 1116 25 37 1369 @® 0 0 0 15 13 169 68 8 64 104 82 20 400 62 2 4 40» Advanced Statistics (B.Com. ~ I1}) (Sem. ~ V1) Nagpur Un ‘| 26 676 728 | any 4 0 | 7 49 al | fa 169 299 Sav=61 | say2= 2633 | Eady = 2900 | Assumed mean of y series = 60 By applying formula : bed — (Sd Sh) p = Rd — Bde Edy ViSae — Gan? Vn. say? aby? Putting the values in above formula ee 10 x_ 2884 ~ (30 x 60 10 x 5488 —(0VI0 x 2584 -(60)2 28840 1800 : V54880- 900 (25840 — 3600 ne 27040 53980 ¥22240 27040 V232336 ¥149.131 _ __= 27040 232.336 x 149.131 27040 34648.500 0.7804 r= 0.78 Co-efficient of correlation = 0.78 r= e Calculation of co-efficient of correlation by log table method ; = Antilog {loz 27040 - a (log 53980 + log 22240) = Antilog {sas20 = ‘ (4.7322 + sar = Antilog {44320 = i oa799} = Antilog { 4.4320 - 4.5397 } 4.4320 97 1.8923 = Antilog { 1.89223 ) A = 0,7798|= 0.7798 +s a is = 0.78 0.7803 Co-efficient of correlation = 0.78‘ ‘Correlation Analysis rr 9 Calculation of Probable Error : Calculation of Standard Error : 1-2 122 PE. = 0.6745 —=- SE. VN 1. (0.782 = 0.6745 = 0.78 Vio = 0.6745 hase ‘ = 0.6745 ae t 4 OSTAS «03016 = 01238 y SE. = 0.1238 = 0.08352597 : * = 0,084 Problem : 2 Calculate the co-efficient of correlation for the following ages of husband & wife Age of Husband || 23 |" 27 | 28 [29 | 30 | 31.) 33 | 35°] 36 | 39 Age of Wife :| 18 22. 23 2) 25 26 28 QO es 0 M32 Also find out Probable Error Solution : A. Calculation of co-efficient of co-rrelation Age of husband = x Age of Wife = y eS dx a2 y dy dy? dxdy P| 25 =F 49 18 ah 495 5 g 49 Hes 27 Sica liso 2 = 9 GE onu 28 =i ea ala ey 4 4 29 = 1 4 a 1 1 ED) Ot ee 0 ® 0 o 0 31 1 1 26 I 1 1 33 3 9 28 3 9 9 35 5 25 29 4 16 20 36 6 36 30 5 5 30 39 9 81 32 i 49 6 {mete | eam | zat=21s | n=10 | 2h=7 | aa2= 163 | xaray— 186 By applying formula : dxdy ~ (Zdv.di mf Vn2de (ear? nz? eae +. rer ‘Advanced Statistics (B.Com. ~ Ill) (Sem. — V1) Nagpur 1), Putting the values in above formula ne 10x 186 = (1x 7. Vio» 215 (VIO « 163-7 ee 1860 ~ 77 2150-121 ¥1630 ~ 49 ee V2029 Vissi 1783 49.044 x 39,762 _ 1783, 1791.04 0.9955 or 0.99 (Approx.) The co-efficient of correlation between the ages of husbands and ages of wives = 0.99 0 * approximately, B. Calculation of Probable Error : 1-2 PLE. = 0.6745 = 0.00424 P.B. = 0.00424] Problem : 3 Find the co-efficient between the heights and weights of 10 students and by test of probable error. Sr. No. of Student Teen fia i CAGES? | 10 Heights (inches) 57 | 59 | 2 | 9 | 4 | 6s [58 | 66 [70] 2 | Weight (Lbs.) u3 | 117 { 126 | 125 | 130 | 128 | 10 | 132 | 140 [149 | Solution : A. Calculation of co-efficient of correlation Denoting Height =x Weight =y x | dr(@4) ae y | a2) ay? aedy | 57 a] 49 13 +8 169 on | 59 a5 25 7 -9 81 45 | 02 -%) 4 ® 0 0 0 63 zi 1 12s a 1 1 | ® 0 0 130 4 16 0 J eeees! a ean 1 128 2 4 2SDL lm Correlation Analysis roe u 88 ~6 36 110 16 256 96 66 2 4 132 6 36 12 70 6 36 140 4 196 84 nR 8 64 149 23 529 184 menOhe) idem 4 Ydr2=220 | n=10 | Ydy=10 | Sdy?= 1288 | Ededy= 515 By applying formula : Bs indy’ ~ (Sd. dy) Vndae ~ (dx)? Vi.2dy2- (ay)? Putting the values in above formula ppt OK SIS NE 4) x 10) "Vio x 220—(4eVi0 x 12880 (10? i 5150+ 40 "* ¥2200- 16 Vi2880 — 100 5190 v2i84 12780 : 5190 46.7333 x 113.0487 5190 «The co-efficient of correlation between x 5283.1388 and y = 0.98 approximately r = 0.9823 or r= 0.98 (Appréx.) B. Calculation of Probable Error : p= PE, = 06745 1== -E = 067457 1-(0.98)2 = 0.6745 vio 1—0.9604 3.1623 0.0396 = 0.6745 3.1623 = 0.0084 P.E. = 0.0084 = 0.6745 Problem : 4 Psychological tests of intelligence of arithmetical ability were applied to 10 children, Here is a record of ungrouped data showing intelligence and arithmetic ratios, Calculate the co-efficient of correlation. ; Child Ra EAUAEB S|, Co] 4D. dame, Me Gel Hy | 1] a Intelligence Ratio: | 105 | 104/102] 101 | 100 | 99 | 98 | 96 | 93 | 92 Arithmetic s_ [101 | 103 | 100] 98 | 95 96 | 104 | 92 | 97 94 2/B. Com.-(Part-I1)-Advanced Statistics (Sem.-V1) Nagpur Uni. itsSolution A 2's Adbvanced Statisties (B.Com. ~ It) (Sem. = VD Nagpur tn, Calculation of co-efflelent of correlation Denoting Intelligence Ratio Arithmetic Ratio = y = dx (99) Fe y [oon | w dvdy | gos ae 101 3 9 | 104 5 25 103 5 25 25 | 102 3 9 100 2 2 is | 101 2 4 98 0 0 Te | 100 1 1 95 3 9 = | ” 0 0 96 “2 4 | 98 -1 1 104 6 36 EO) 96 a9 9 2 -6 36 18 93 -6 36 97 -1 1 6 92 -7 49 94 -4 16 28 | n=10 | Sae=0 | sata | n=10 | zayc0 | za?= 140 | Bdrdy- 92 | By applying the formul n.Bdxdy— (Sdr.Bdy) Vn. de? — Bar? Vn 2a? hy? Putting the values in above formula 10 x 92~ (0x0) Vio x 170-(0pVi0 x 140-(0)2 920 Vi700 ¥i400 a 920 © 41.2310 x 37.4166 920 1542.7238. 0.5963 or r = 0.60 (Approx.) r= W r Problem : 5 The following table gives the marks obtained by a group of 12 students in two examinations A & B, Calculate the co-efficient of correlation between the marks obtained in the {wo examinations. {Student ee | ae Marks i 13 [17 | 14 | 18] 12 [20 [16 fas faz fas | 2 Marks infixam B : | 18| 16 | 18 | 15 | 19| 16 | 18 | 15 | 21 | a7 | 18 | 20 Solution : Calculation of co-efficient of correlation Denoting Marks in Exam A. =x Marks in Exam B = yx | «an at et a 4 WW 4 16 v 0 0 4 ’ 9 8 1 1 n $ 25 16 ~2 4 10 2 3 9 18 0 0 0 16 et 1 15 -3 9 3 Is 1 1 21 3 9 3 7 0 0 17 -1 1 0 19 2 4 18 0 0 0 21 4 16 20 2 4 8 m=12 | Yde=-5 | ya =91 n=12 | Xdy=-5 | Yd= 41 | Ldedy= 42 By applying the formula : 4 n.Bdedy — (Bde Sdy) Vidar? — (Bae? Vn. 2dr? — Baye Putting the values in above formula be 12x 42~ (-5x-5 Vi2 x 91-5 Vi2x 41-(-5pP a 504-25 Vi092=25 V492 - 25 pe 479 "* Vi067 Va67 479) V¥32,66410 x 21.6102 Ser "© 7058777 r = 0.6786 or r= 0.68 (Approx.) 4 r Problem : 6 Calculate the co-efficient of correlation between the test scores and sales TestScores :| 14 | 19 | 24 | 21 26 Da). 015 Sales (100%) | 31] 36 SUR | Sees! (39 | Solution : . Calculation of co-efficient of correlation Denoting Test Scores = x Sales = ypss Advanced Staristics (B.Com. = Il) (Sem - VD Nagprp yy, y | von | 4 | ri 8 64 43) 46 ae 9 3 M4 4 16 48 9 4 | s2 1 \ 9 2 4 ~2 % 6 36 50 ul iN : 22 2 4 4s 6 36 Pa | Is aby as 3 NG 36 0 0 0 4 2 ‘ a 19 a 1 ® 0 0 0 | m9 | Sdv=0 | a2=120 | n=9 | say=9 | Ed?=355 | Pdrcy = 105) By applying the formula : n.Ddrdy ~ (Rx Edy) Vin Bar? (Bax)? Vn.2h? ~ (Say)? Putting the values in above formula uit 9x 193 ~ (0x 9) Vo x 120 ~ 7 Vo x 355 - (9) wee 1737-0 i080 00 V3195 81 eebeia ST i080 V3114 i pee eet LIST. ANE 32.8634 x 55.8032 1737 7 183.8829 r = 0,9472 or r = 0.95 (Approx.) Problem : 7 Find out the co-efficient of correlation between the heights of fathers and sons. Heights of fatheres (inches): | 65 | 66 | 67 | 67 | 68 | 69 |) 71 | B | Heights of sons inches): | 67 | 68 | 64 | 68 | 2 | 70 [ 6 | 7 | Solution : Calculation of co-efficient of correlation Denoting : Height of father =x Height of Son = y x dx (68) a? yi dy (69) a? dxdy 65 -3 9 67 2 4 6 66 -2 4 68 = 1 2 67 1 64 -5 25 5 67 1 68 =1Putting the values in above formula 2 8x20- 2x0 a Vs x 50-2 V8 x 42- (07 ey 160 +0 Va00—4 336-0 160 396 336 cS 168 19.8997 «18.3303 2 Soe, 364.768 r = 0.4386 or r = 0.44 (Approx.) Problem : 8 Calculate co-efficient of correlation between advertisement cost and sales as per the data given below. Advertisement Cost (00.000) :| 39 | 65 | 62 [90 | 82 | 75 | 25 | 98 | 36 | 78 ~ Sales (00,000) 2] 47 [53 [58 [86 | 62 | 68 | 60 ou [st | Solution : Calculation of co-efficient of correlation Denoting : Cost = x Sales = y x dx (65) dx? y dy (62) dy dxdy 39. | 226 676 7 iy 2s 390 0 0 3 9 81 0 wa 9 58 -4 16 R 25 625 86 4 576 600 ” 289 ® 0 0 0 10 100 68 6 36 60 40 1600 0 La 4 80 3 1089 91 29 Bal 957 -29 41 51 “ul 21 319 13 169 84 2 484. 286 Faaie| eatmamal]y neta | wpa [apts [nassom, = Ht) (Sem VD Noxpur e Advanced Starstion (8 Ry applying the formula : 7 Tyee = (at? yout “adv? Putting the values in above formula 7 Troe non cay? Vio 2384 = (40)? re Tena emrese soso V23840 = 1600 pe et Vss080 V22240 Bie nek 27D 332336x 49.131 pe 2TH 34648.500 r= 0.7804 or r = 0.78 (Approx.) Problem : 9 Calculate co-efficient of correlation between Price & Supply s Pie @:] 17] 18] 19] 20] 21 | 2 | 2 | of | ee | Supply ®_38_| 37 |* 33 [ean csmeetaieee T) 26 | 23 Solution : Calculation of co-efficient of correlation Denoting : Price = x Supply = ¥ x | aay ae vy | ey ae dedy "7 a 4 38 ne 81 m8 tert 1 37 8 64 -8 19 0 0 38 9 8 0 20 1 3 4 16 4 21 h 4 32 3 9 6 mis] 713 9 33 4 16 2 2 4 16 34 5 Sia 20 24 5 25 29 0 0 0 25 6 36 2% 1 9 1 [mmo | av=25 | a?= 145 By applying the formula ; Inte an dy ~ (dy)? osCorrehacion Anatysis o's” Putting the values in above formula re 10 x44 Sx 33 Vio x 145 oes 337 = (33)? 440 ~ 825 i. Via40 625 3370 1089 1265 r 825 V2281 a’ = 1265 28.7228 x 47.7598 _ = 1265 "* 7371.7952 r = -0.9221 or r= ~0.92 (Approx.) Problem : 10 Calculate Karl Pearsons co-efficient of correlation from the data given below : xp Se 2 | 4 | 6 | 8 [eroto es ips Sa moe 14 16 | 18 pikca0 Solution : Calculation of co-efficient of correlation “2 dx (6) ae y dy (16) dy? dxdy 2 -4 16 12 a 16 16 4 = 4 14 -2 4 4 ® 0 0 ® 0 0 0 8 bi 4 18 2 4 4 10 4 16 20 4 16 16 n=5 | Xde=0 | 3cr?=40 n=5 | Sdy=0 | Zah2=40 | Ydxdy=40 By applying the formula : if n.duedy — (Ede. TS Mnzde - ar? Vn 2a? - a)? Putting the values in above formula “ 5x40~ (0x0) TONS x 40—(02V5 x 40-(02 _ 200 ”* 200 x ¥200 id 200 14,1421 x 14,1421 a. 199,99900 = 1,0000 or = 1.0 (Approx.) es a.ww anced Suatietics (BCom « Ml) Bem Vi) Nay, Pasting the values in above formula is Sx a) Srey ie 40 (0) . re 200 | 200 x ¥200 200 14.1421 9 14.1421 . —=200 "* 799.9810 r= ~1,0000 or 7 = ~1.0 (Approx) Problem : 11 Calculate co-efficient of correlation of the following figures relating to the consump, Wzers and the output of foodgrain in district ‘X” |___ Chemical feritizer used : ] 100 no | 120 | 30 | 140 | 1507 is I fertili (in matric tons) : | 1000 | 1050 | 1080 | 1150 | 1200 | 1220 Output of food =| 170 | 180 | 199 | 200 | 210 | 220 (in matric tons) 1360 | 1420 | 1500 | 1600 | 1650 | 1650 Solution : Calculation of co-efficient of correlation | Denoting : Chemical fertilizer used = x 2 Output of food = y Note : The figures of the both the series divided by 10 for the calculation purposes. a [= [aay] aa y | #039 | a? dxdy 10 Ls 25 100 ~36 1296 130 Nl 24 E16 105 -31 961 124 12 3 9 108 Es 784 84 13 = 4 115 -21 441 42 . 14 = 1 120 -~16 256 16 ® 0 0 122 -14 196 v0 16 1 1 130 ~6 36 6 17 2 4 G& 0 0 0 | 18 3 9 142 6 36 16 19 4 16 150 14 196 56 20 5 5 160 4 576 120 2 6 36 165 2» 841 174 2 I 49 165 29 841 203 B 8 64 165 29 841 232 n=l4Corretanion tatysis vor By applying the formula : 5 ydavaly «(Sade ddy) Vinsae (Sat)? Vy Ley? = ay)? Putting the values in above formula .. Jax 1243 x-21 vi x 289 (a1 Vidx 7301 (21 pw ome 1402 + 441 V3026 441 x Vi02214 441 17843 3185 x V101773 17843 136.436 x 319.0188 _ — 17843 18004.145 r= 099 Problem : 12 Calculate the co-efficient of correlation from the data given below. wee. 1200 1000 -800 400 1200 1400 | -600 | 1000 | Y_: | -3000 | 3300 | 2400 | 1200 | -3600 | 2100 | 1800 | 3000 Solution : Calculation of co-efficient of correlation. The figures of the both the series divided by 100 for the calculation purposes. x dx (0) ae ’ dy 0) dy? dxdy | 2 2 Tamm Weaopuhl 36 1296 = 432 | tip, je 10 100 33 33 109 330 | -8 -8 64 24 24 576 -192 -4 -4 16 12 12 144 -48 12 2 1444 5]'=36.4|'' -36 1296 =432 14 14 196 -21 -21 441 294 a * 36 18 18 324 = 108 Bio Ae) 100 30 30 900 300 | e-5 Zar=0 | Za2=800 | n=8 | Sdy=24 | ra2= 066 | xexdy= 2136 By applying the formula : ee n.Bdidy (Bdx Day) fy. Ba? — (Bax)? Vin ay? — (ay)? Putting the values in above formula a 8x 2136 ~ (0x24) 1B x 800 ~ (0)? V8 x 6066 ~ (24)? ~ 17088 r= Ves00 x Va8s28 ~ 5761 = Ml) (Sem. - VI Nog, Advanced Statisties (B.Com. PUP Up rrr =17088 te Ver0o x Vav0s2 17088 80 x 218.9795 = Voss VISI8.36 r= -0.9754 or P= =0.98 Problem : 13 he flowing table gives the results of matriculation examination eld in 2018 aie Kay Pearson's coefficient of comelation and its probable error. Do you think age is related Perventage of failures? : AgeofStudents | Percentage | _Age of Students a : (years) | _offaitures (ence) Ores ) B-14 39 1 fa 14-15 | 40 19-20 ye 15-16 IS 20-21 ta 16-17 B 21-22 56 17-18 , 36 — Solution : A. Calculation of Co-efficient of Correlation L | myx | d(175)| a2 y | dy (3) ay dxdy | ns tase 16 390 (4 16 +16 | PR: 9 Hoa |S 9 +9 ISSiteli hese, 4 4B 0 0 0 165 =I 1 0 0 0 @ 0 0 Bosh 7 49 0 | 185 1 1 Bodie 4 16 -4 aed 195 2 4 4 | +5 25 10 | 205 3 9 bab a 1 3 215 4 16 56 +43 169 52 n=9 | Xde=0 | Bax?=60 | n=9 | Bay=+1 |ea2=285| Saray 86 By applying the formula : 4 1 2drdy ~ (Sdr.Zdy) ” Vazae = Gan? Vn.2ay?— (Bay? Putting the values in above formula a 9x 86~ (0x1) [hoe 60-0) V9 x 285 — (HP SL aaan0) " * W540—0 x V2565—1 774-0 "* V5a0-0 2565-1Correlation Analysis 774 Vs40 x 2564 774 pocnya 176.679 P= 0.657834 r = 0.658 ~ 23.338 x 50.636 B. Calculation of Probable Error : PE = 0.6745 4=10.658" vo - 1 ~ (0.432964) 0.6745, 3 = 0.6745 = 0.6745 P.E. = 0.1274885 0.567036 3. PB 0. rar 0.6745 x 0.567036 3 1274885 TYPE NO. 2) uy Note : If three series are given in the Problem, which are to be converted into two series on a particular basis. Problem : 14 The following tables gives the classification of population and death by heart disease among them. Find out if there is any relation between the age and deaths : Age: 0-10 10-20 | 20-30 | 30-40 | 40-50 | 50-60 | 60-70 | 70-80 Number of Persons : so | 45 | 30 25 | 10 lg Ba 2 (In thousands) Death : A0m [paar {ese [iro [diaz-,) |nllaoe | oa 0. Solution : Being the Correlation is to be calculated between age and death Therefore > Age=x Death = yin thousand ‘The death are out ofthe population (No. of persons) and no, of persons are different for each group, means the No. of death are on the differnt basis. To take the deaths on a particular base, we have taken the base 100 thousands. Total Population | Deaths Equation Death in 100 (in thousands) thousands ‘y’ series 50 40 50: 100 :: 40 _ 100 40 80 ra so 45 27 45: 100::27 _ 100x 27 60 « BS| vere Advanced Statistics (B.Com. te oe W Nagy, 7 36 30; 100 1:36 ~ mast I29 Po ee | Bt x 40 25: 100 1:40 glen 4o 160 Le aes ee | 18 10 18 100: 10 _ 100%18 180 10 7 20 7: 100 ::20 10x20 286 3 10 3: 100::10 - ext 333 a 10 2;100::10 plovx)0 500 ie 2 Calculation of co-efficient of correlation : r Age | x |dr@sy] ae y | dy (180) ay dxdy (m.y.) 0-10 5) a) 900 80 | -100 10000 3000 10-20 | 15 | -20 400 60 | -120 14400 2400 20-30 | 25 | -i0 100 120 | -60 3600 600 3040 35 0 0 160 -20 400 0 40-50 45 10 100 180 0 0 0 50-60 | 55 20 400 286 | 106 11236 2120 60-70 65 30 900 333 153 23409 4590 70-80 | 75 40 1600 soo | 320 | 102400 12800 n=8 | Xdv=40| Zax? =4400 | n=8 | Zdy=279 | Zdy? = 165445 | Zdrdy = 25 By applying the formula : Batrdy — (Sd 3dy) Nn.Eax? — (Eds)? Vin2ah? — Bah? Putting the values in above formula * fas 8x 25510— (40 x 279) V8 x 4400 - (40)? V8 x 165445 — 2792 i 204080 — 11160 ” ~ 35200-1600 1323560 - 77841 _ 192920 "33600 Vi245719. 192920 * 192920, ~ 204587,7411 r = 0.9429 0.943 ~ 183.3030 x 116.1178 Co-efficient of correlation = 0.94Correlation Amalsis aor wy Calculation of correlation by log table method : 192920 r= ¥a3600 x V 1245719 By applying log table = Antilog [ioe 192920, tog 33600 + log 1245715) = Antilog [sasss - 514.5263 + 6.0955) 1 = Antilog [s28s3 2 (10.218)] = Antilog [ 5.2853 - 5.3109] = Antilog 1.9744 0.9428 0.943 Co-efficient of co-rrelation = 0.94 B. Calculation of Probable Error : Ine P.E. = 0.6745 Vn = oe74s 110.943)? 4 v8 = 0.6745 1~ (0.889249) 2.8284 0.110751 = 0.6745 2.8284 PE. = 0.0264 Problem : 15 The following tables gives the distribution of the total population-and those who are wholy or partly blind among them. Find out if there is any relation between the age and Blindness : Ages No. of Person’s Blind (in thousand) * 0-10 | 100 TSN Auge t ee 10-20 ¥ 60. , 40 20-30 40 40 30-40 36 40 40-50 24 36 50-60 ° ul 2 60-70 6 18 70-80 3 15 Solution ; Being the Correlation is to be calculated between age and blindness. Therefore ~ Age=x Blindness = y The blindness are out of the population (No. of persons) and no. of persons are differnt for ¢ach group, means the No. of blindness are on the differnt base. To take the blindness on a particular base, we have taken the base 100 thousandes.. peed Marisier (0.60 110 FBO VE Hig, ineeeereenpeT Winds ons 44 100) Mournny, “Vso hey 14000 2 iat a Bai ‘By applying the formula : ab neh in Br? ~ (Lade)? Vy. Edy? ~ (Zdy)? Putting the values in above formula As Sx 248102) 40 283) 8 x 4400 ~ (40)? V8 x 167435 ~ (283)2 "2 Fis200~ 1600 Vi339480~ nooKoCorrelation Analysis vor 7 184760 33600 x ¥1259391 184760 183.3030 x 1122.2259 184760. 205707.3741 r= 0.8982 r = 0.898 «. Co-efficient of correlation = 0.898 = .90 Calculation of Probable Error : 1 ~ re - re Vn = 2 0.6745 += O898" V8 1~ (0.806404 0.6745 2.8284 0.193596 2.8284 _ 0.6745 x 0.193596 R 2.8284 _ 0.1305805 ~ 2.8284 PE. u 0.6745, u 0 = 0.6745 P.E. = 0.0461676 P.E. = 0.046 Calculation of correlation by log table method : oe 184760 ¥33600 x 1259391 By applying log table = Antilog [oe 184760 -} (log 33600 + log 1259391 J = Antilog [52667 = (4.5263 + 6.1000) = Antilog [52667 = 5 (103263)| = Antilog [ 5.2667 - 5.31315 ] =. Antilog [ 7.95355] 0.8986 0.898 Co-efficient of co-rrelation = 0.90 Problem : 16 The following are the results of B. Com. examinations Ages Candidates Candidates of Candidates Apperaed Successful 13-14 200 | 124 ~ 1445 300 180 15-16 100 65 “. wre Advanced Maiates (Com, = 10) Bem. VY Sor Ww? oe a aa vd ihei9 a 19-20 e 20-21 31 21-22 2 22-23 15 2 Calculate co-efficient of corelation between age and successful eandidates. Solution : Being the Correlation isto be calculated between age and successful Therefore = Age =x pata? The successfuls are out of the total candidates appeared and candidates appeared arc « ‘or each age group, means the successful candidates are on the different Bases. For the calculation of correlation the data must be on a perticular base. ___ Therefore we have taken the base of 100 & prepared 'y’ series as given below : Candidates | Candidates Equation Successfu appeared Successful Candidates in Yay 9" series piss 2 si 200 124 200:100::124 | _ 100x124 2 ' 200 300 180 300: 100::180 | 100x180 60 F300. Su Leh ae 100 65 100 : 100 :: 65 _ 100 x65 65 ~~ 100 50 34 50: 100 2:34 10034 68 aR — | 4 150 99 150: 100 :: 99 _ 100x99 66 Sul. ~ 150 400 252 400 : 100 :: 252 100 x 252 63 400 250 145 250:100::145 | 100 x 145 58 5) 250. 150 81 150; 100 :: 81 100 x 81 54 150 25 2 25: 100 :: 12 100 x 12 48 ce 5 18 3 75100 ::33 _, 100 «33 44 onsCorretion Aa ver Calculation of co-efficient of correlation | x(a) « |» loo] dtedy 13s Pe (hate at 1d 16 -16 las \ 9 o | 2 4 -6 Iss 2 4 ef 49 -14 ws | -1 1 68 | 10 100 =10 GD] o 0 66 8 64 0 Tesh 1 6-| 5 25 5 19.5 2 4 0 9 0 205 4 9 s4 | 4 16 ai 215 4 16 48 | -10 100 -40 ns 5 25 44 | -14 196 =70 | | m=10 | xar=5 | sa2=as | n=10 | xay=e | sa2=570 | Saedy=—163 By applying the formula : nddrdy ~ (Zdx. Edy) Vn.dde = (Bak)? Vn. day? — (Bay? Putting the values in above formula ae 10x -163 (5x8) Vio x 85 — (5)? V10 x 570 - (8) 1630 — 40 850-25 5700-64 — 1670 825 x ¥5636 3 ~ 1670 1 * 28.7228 x 75.0733 Bs 1670) 1 © 21563154 r . cr = (0.775 . Co-efficient of correlation = .78 Problem : 17 Calculate co-efficient of correlation between age and Playing habit from the following data : Age Population No. of Players 15-20 1500 1200 20-25 2000 1560 25-30 4000 2280 30-35 . 3000 1500 35-40 2500 1000 40-45 1000 300 45-50 800 200 50-35 500 50 55-60 200 6 3/B. Com.-(Part-II1)-Advanced Statistics (Sem.~VI) Nagpur Uni,- = IIl) (Sem. ~ Vi) n, a idvanced Statistics (B.Com. PNaging rr a n Solution erie. ot, i abit (No. of Play "ein the Correlation isto be ealeulated between age and Playing D>”. © i Therefore (2) Playing habit (Amone No- of players) (1) Age =x tion are differnt for each age Ey Hy "he No. players are out of the population and popula Mens the ditfemt no, of players are on differant bases, 4180 "OF the correlation the data must be on a pericular base, Thorefore we have taken the base of 100 & prepared *y’ series eae oc Population No. of Equation >) ei fPlayers in - 5 'Y’ Series i Players : 1500 1200 1500; 100 :: 1200 | _ 100 x 1200 80 = "1500 2000 | 1560 2000; 1001560 | _ 100x 1560 eB 2000 4000 2280 4000: 100 2280 | _ 100x 2280 57 4000 3000. | 1500 3000: 100:: 1500 | _ 100 x 1500 | = “3000 2500 1000 2500 :100:: 1000 | _ 100 x 1000 40 = "2500 1000 300 1000: 100::300 | _ 100x300 a = 1000 800 200 800: 100::200 | _ 100x200 25 ~~ 800 500 50 500 ; 100 :: 50 _ 10050 10 To600 200 6 200 : 100 ::6 100 x 6 3 200 Caleulation of co-efficient of correlation : x dx (37.5) dx a dy (40) dy? dxdy 175 | seas 400 80 40 1600 ~ 800 728s 225 ns 38 1444 570 208 10 100 57 7 289 170 32.5 6 25 50 10 100 -50 Gp 0 0 0 0 0 425 5 a5 30 -10 100 =50 475 10 100 25 eis 25 ~150 52.5 15 25 10 |e 3g 900 450 pie 20 400 i -37 1369 -740 n=9 | xdv=0 | av=1500 | n=9 | say=13 | a2= e007 | Say = 2080Coercion Anat By applying the formula ; nde nddedy 2~ (Yds) Putting the values in above formula ta 9 x 2080 = (0x13 Vox 1500 (0)? V9 x 6027 aay 26820 - 0 in Viss00 5453 169 , r r é Vi3500 26820 ~ 0 Vsa074 26820 © 116.1895 x 232.5382 = 26820 27018.4972 ) 0.9926 Co-efficient of correlation ddy? = (dy)? Problem : 18 Find out correlation co-efficient between age and Playing habit of the following students : Age in Years mi 5 16 {imap 18 19 20 Number of Students : | 250 200 150 120 100 80 Regular Player 200 | 150 90 | 48 30 12 Solution : Being the Correlation is to be calculated between age and playing habit (No. of Players) (2) Playing habit (No. of players) =y Regular players are from the No. of students and No. of students are differnt for each age group, means the regular players are on different bases. For the correlation the data must be on a perticular base. Therefore we have taken the base of 100 & prepared ‘y’ series Therefore (1) Age =x No.of | Regular F No. of Players in Students | Player Eyiaton %, ‘y’ series ; ~ 100 x 200 250 200 250: 100 :: 200 Piao 80 100 x 150 | 200 150 200 : 100 :: 150 Ra 15 100 x 90 150 90 150 : 100 :: 90 aca 60 100 x 48 120 48 120: 100 :: 48 = x 40 100 x 30 one = 10s 30 30 100 30 100 : 100 ::30 a eel is 10012 bs ,* R 80; 100 1:12 aa |Putting the values in above formula - E 6x 210 ~(3 x60) a Vex 19— a (oP = 1260 — ~ Viie— seer ce = 1440 © Wi0s x-V20100 = 1440 ~ 10.2469 = 141.7745 = 1440 1452.7491 her 099 + Co-efficient of correlation = — 0.99 r= Problem : 19 From the following data find out if there is any realationship between density of popula:io: and death-rate. District | Areain aS. kms. | Population | __No. of Death 1 jet 7,00,000 14,000 2 | 00 6,60,000 10,560 3 2,40,000 4,320 4 a | 1,80,000 3,060 5 900 | 2,70,000 5,130 6 500 2,00,000 3,000 Solution : Being the Correlation isto be calculated between density of population & death - rate. Therefore — Density of population = x Death-rate = yCoretain Aral sts err i Aut in this problem density of population & death rate both are not given therefore first of all we have to calculate density of population (x Series) and death rate (y Series) as per below. Y Series Calculation of density of population + Calculation of Death - Rate + (Density of population means population in | (Death - Rate means Death in every 1000 Persons) ‘one square Km.) _. Total Population Density = “Total Sq. Km. 2,00,000 1. 7,00,000 : 1000 :: 14,000 e190 ee , 1,000 14,000 _ 49 7,00,000 > 660,000 = 600 2. 6,60,000 : 1000 :: 10,560 As1p0 1,000 x10,560_ 6,60,000 2.40,000 3, 2,40,000 : 1000 :: 4,320 3. 300 = 300 1,000 x 4,320 Le 1g 2,40,000 1,80,000 4. 1,80,000 : 1000 :: 3,060 Oe salah on 1,000 «3,050 _ 1,80,000 : x 5,130 270.000 eee 5. 2,70,000 : 1000 900 1,000x 5,130 _ 4g 2,70,000 6. 2,00,000 : 1000 :: 3000 6g, 7:00.000 4g 500 1,000 x3,000 2,00,000 Calculation of co-efficient of correlation : x dx (400) ae y dy (17) ay? dxdy 700 300 90000 20 5 9 900 600 200 40000 16 -1 1 200 300 | -100 10000 18 1 1 — 100 300 — 100 10000 7 0 0 0 300 | -100 10000 19 2 4 200 400 0 0 15 -2 4 0 n=6 | Sdr=200 | Zax?=1,60,000 | n=6 | Bd=3 | Ed?=19 Ededy =— 400 By applying the formula : n.Zdxdy — (Sdx.Zdy} 7S Vnzae2— ds)? Vn ay? - (Bay?Com =I Sem — Vy, R Advanced Starts rer Putting the vahues in above formula * 6x 400 ~ 200 Vox 160000 (200)? Sr Y 2400 — 600 Voeoooo = ab000 Vite r8 my 1800 920000 x V105 1800 a ae 959.1663 x 10.2469 1800 mn "* 984812 r= 0.1831 r= Q18 Co-efficient of correlation = 0.18 (CO-EFFICIENT CORRELATION BY TWO WAY TABLE A Bivariate Distributi ‘Wo variables are paired distribution data, the m ion refers to data conceming two variables ‘x’ and ‘y’ 7 a this ¢: and then their coefficient of correlation is calculated. in : bivariate tumber of pairs of observation is large then it becomes very lic calculate the co-efficient of correlation in such a large pairs of observation, the data are su in the form of a “Two Way frequency table”. It is aso called “Bivariate frequency Tabi. Contigency table or correlation table. Ina two way freugency table of two variables x and y, the class interval of x series the first row ontheT: ‘opofthetableandtheclass interval of second series y are put in the firs the left of the table - In this type formula of co-rrelation will be as per below : st u.Efidedy — (Sf 3fy) Vase — 3785) Vn? a r = correlation 1 = Total of frequency fax =£@ i oR OX & Sty =Q X i OR Qx w Jas? (fx) = fex @ i OR fae Q ax JO? (ty) = fy @ i OR fity @ 4 faedy =f X1 Qi oR Ox @x w& Problem : 20 : Calculate the co-efficient of correlation for the following data : ] Age of Age of Wife “| husband 19:20/7 4 20-90 | ancag gies 50-60 Total 15-25 6 2, ec - _ 9 25-35 3 1s in Fs 7 iCorredation Amadysis: b oro 38-45 its ‘sq i 14 45-55 E 6 55-65 hake | oa ale 31 welaws Solution A. Calculation of co-efficient of correlation By applying the formula : ” Bfitedy — (Bfete.Bfey © Ni. 3p? — ft? Vn 20? — Bs? Putting the values in above formula 100 x 99 (~7x -8 © Vi00 x 123-7)? Vi00 x 122-8? a 9900 — 56 ~ ¥12300— 49 ¥12200 - 64 + 9844 © Vi12251 x ¥12136 9844 110.6842 x 110.1635 9844 12193.3589 r = 0.8073 r= r= “r= 0,81 (appro.) B. Calculation of Probable Error 2 P.E. = 0.6745 a n™ Cm = MD (Se y Aatvormrerd Starntien 1 MD Now, ~ aera LQ? 10) = cers & Wwasgu) 0 * ners MM o ~ ROT «094s Ly - GES te Prediem : 21 be “slculave the co-efficient of correlation between the marks obtained by a batch of jo _Scontancy and Stastics in the following table, Stasis in the flo een Marks in ] ia. Marks in AccountancyCorrebanion Analysis ovr By applying the formula ; i 1 3iledy = (fd. 3/y) Vnyite~ (37k)? Vn Sev —yapye Putting the values in above formula 88 (-) (Rx = 34 — Pee Vioox 92 ~(8)? V100 x 154 ~ (34) 8800 +272 9200 64 -V15400 ~ 1156 9072 19136 x Via2aa c 9072 95.5824 x 119.3482 = 0.80 (approx.) = Problem : 22 Find out the correlation from the following data. [x —¥| 30-35 | 3540 | 40-45 | 45-50 | 50-55 | 55-60 | Total 80-90 2 3 2 th ee Hi q 90-100 s 2 5 4 2 - 13 100-110 = 4 8 5 1 Ei 18 110-120 a ng 2 3 1 1 ov 120-130 1 = = 2 1 1 5 Total 3 9 7 14 5 2 50 Solution : Calculation of co-efficient of correlation. a 30-35] 5-40] 40-46] 45 60]60-5566-60 Total] M.V.| 92.5 | 37.5] 42.5) 47.5 | 525|57.5|fad roe ‘Advanced Statistics (B.Com. ~ Ul) (Sem. VI) Naypyy BY applying the formula : | i Nerney (Xftey? ae = (fy? Putting the vatues in above formula | be 50x 26 ~ (1S x —10) Vox 73=(1sy ann Caee Ph 1300_+ 150 V3650 ~225 3400 — 100 _ —_14s0 V3425 x ¥3300 " 1450 58.5235 x 57.4456 5) 33619176. r = 0.4313 r - 0.43 Problem : 23 Calculate the Co-efficient of Correlation. Y «x? Series E Series 16-18 18-20 20-22 22-24 Total 10-20 2 I I S 4 20-30 3 p, 3 2 10 30-40 3 4 5 6 18 40-50 2 2 3 4 ul 50-60 - 1 2 a 5 60-70 - 1 2 1 4 Total 10 Nl 16 15 32 Solution : Calculation of co-efficient of correlation.Corrs Anadis 7 By applying the formula : oY nZfdrdy ~ (Zfdx.Zfdy) Vn 2/2 — (fs)? Vn. 32 — Cy? Putting the values in above formula a 52 x_16 = (-16 x 15 52x 66 - (— 16)? vs2 x 93 —(15)2 832_+ 240 3432 — 256 V4836 — 225 eae) 072 ~ 3176 x V46i1 3 1072 56,3560 x 67.9043 1072 ” * 3826.8172 r = 0.2801 r = 0.28 (app.) r = 0.28 Problem : 24 Calculate the co-efficient of correlation. Marks in Marks in Accountancy Total |__ Statistics 20-25 25-30 30-35 35-40 40-45 15-25 40 20 4 = - 64 25-35 8 48 12 4 = n 35-45 - 10 2 8 4 44 45-55 - 10 2 4 2 18 55-65 - - 2 - 3 4 Total 48 88 42 16 8 202i wee Avon Sates (B.Com 11) Bem yy ay Solution ; Catoutati 7 alculation of co-efficient of correlation [afal otal Pi bd lal" ce By applying the formula : Sfiixdy — (Bfax.¥, T Mnape— fe Vn 3b Ra Putting the values in above formula p = 202 266 = (152 x- 174) V202x 328 — © 152)? V202« 362 ie 53732 26448 166256 - 23104 73124 - 30276 Be tea 2728A 7 5743152 «42848 S 27284 ~ 207.7306%206.9976 _ 27284 1 * 42999.7356 r = 0.6345 (app.) r = 0.63 Problem : 25 Calculate the co-efficient of correlation between the total cultivate area and the area un wheat. r Area under Total Cultivable Area | Wheat 0 500 1000 1500 | 2000-2500 | Tots! 0 18 9 - & BS 2 200 3 2 6 4 2 2 400 - 6 10 4 " 20 600 Es 2 - 5 1 6 800-1000 a 4 = 3 2 SU Total 21 44 16 14 5 100Croat Anat ore » Solution; Calculation of co-efficient of corre! lation |—>Shidy -132 By applying the formula : ht Diddy — (Bfetx.Bfdy Vn. — (Brey? Vn 34? — Cy? Putting the values in above formula Bs 100 x_132 — (~62 x - 80) Vi00 x 162 —(-62)2 ¥100 x 176 — (80)2 E: 13200-4960 ~ ¥16200 = 3844 117600 — 6400 it 8240 © ¥12356 x ¥11200 Ba 8240 111.1575 x 105.8301 8240 ” * 11763.8093 r = 0.70 (app.) r= 0.70 Problem : 26 Calculate the co-efficient of correlation and the probable error Weight Age in years (Kg) 20 22 24 26 Total 100-105 20 15 10 3 48 105-110 10 ist ak 5 2 110-115 9 R + 9 30 115-120 1 7 6 7 4 120-125 E 6 e i 6 Total 40 40 16 24 120F 4 : or Advanced Starstics (B.Com — ID (Sem.— yyy. Solution ; Calculation of co-efficient of correlation, By applying the formula : ». n.Sfdedy — (Sfx Day) Vn Byte? (fee)? Vn Bf? — Bfey?? Putting the values in above formula ae 120 x 27 — (24x -92 i120 x 152 — (24)? Vi120 x 252 — (92) 3240 — 2208 gk eS 32A0T 2208 nee V18240 — 576 V30240 — 8464 Pipe bik Saabs 17664 x 21776 ne 5448 132.90510 x 147.5669 _ 5448 r © 19612.3936 r = 0.2778 (app.) r = 0.28 Problem : 27 Find the co-efficient of correlation between age and the sum assuared from the follc table. if . Age Sum Assured , Group . 500 . 1000 2000 5000 40000 | To 15-24 18 20 6 2 f 4 25-34 21 26 6 5 1 35-44 10 9 3 6 |Coretation Amaya vee 4“ "45-84 Ly 7 5 a4 88-64 8 3 1 12 [Tow [64 66 at 170 Solution : Calculation of co-efficient of correlation, By applying the formula : Bs n.Bfdedy — (Bf: Zfay) Vin. 3fa — (2k? Vn. 2h? — ty)? Putting the values in above formula J 170 x 160 ~ (190 x ~103) 7 Vir x 1964— ©1907 Vi70 «315 = 103)? sy 27200 + 19570 ” © 7333880 — 36100 V53550 — 10609 i 7630 "© ¥297780 x V42941 7630 ” © 545,692 x 207.2221 1630 ” * 113079.4836 r = 0.0675 (app.) r = 0.067 RANK CO-RRELATION (R) Steps : 1. Give the ranking in ‘x’ series from higher value to lower value and denote it by Ry. 2. Repeat the same process in 'y’ series and denote it by Ra. 3. Ifthe values are repeated then "Average Rank" should be calculated. nee