Scribd
Upload
116 views28 pages

M 4

mathematics 4 unit 4

Uploaded by

Kaleen Bhaiya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
116 views28 pages

M 4

mathematics 4 unit 4

Uploaded by

Kaleen Bhaiya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 28
Statistica] Techniques | CONTENTS — Part-1 Part-2 Part-3 Part-4 Introduction : Measures of 3-2C to 3-8C Central Tendency, Moments, Moment Generating Function (MGF), Skewness, Kurtosis Curve Fitting, Method of Least. e Square, Fitting of Straight Lines; Fitting of Second Degree Aer Parabola, Exponential Curve | { | | Correlation and Rank Correlation......3-15C to 3-18C Regression Analysis : 1. S-18C to 3-23C Regression Lines of y on x and.x on y, 2 Regression Coefficients, < Properties of Regressions Coefficient and Non-Linear Bes easion 3-2 (CC-Sem-3 & 4) Introduction : Men i gees : Measures of Centi deny, Moment Moment Generating Function (MOR) Shanon en Statistical Techniques I the origin. ay? 7 —(-1.5)? = 14.75 3 py'Hy + uy — 80) — 3 (— 1.5)(17) + 2 (- 1.5)8 = 39.75 big = Ha —4 yg’ +6 (Hy 7 (Hg) 30H = 108 - 4(- 1.5)(— 30) + 6(- 1.517) ~ 3(- 1.5)* 42.3125 Moments about origin, 4-15-25 i + (04)? = 14.75 + (2.5) = 21 = Hy + 30,v.—20,° 19.75 +3 x 2.5 x 21-2 x (2.5) = 166 = 60,20, + 3044 5 x 166 — 6 x (2.5)? x 21 + 3 x (2.5)¢ ut Avyvs 49.3125 + 4 x 2. 132 Define skewness and kurtosis of a distribution. The are 0, 2.5, 0.7, and 18.71. Find eee first four moments of a distribution skewness and kurtosis. the coefficient o! wwness means lack of symmetry i., when 8 ‘henit is called a skewed distribution and this The term ske' d or negatively skewed. it symmetric be positively skewer Skewness distribution is no distribution may Mathematics -IV 3-3C (CC-Sem-3 & 4) Kartosis It tells whether the distribution, ifplotted on @ graph would give Kurtowmal curve, acurve more lat than the normal curve, oF more peaked than the normal curve. Numerical: 1hy= Oy y= 25 ty = OT y= 18-7 : ng _ 0.7? Coefficient of skewness (B,) = 85 = = 0.08136 (+ ve) 2 25° ‘The distribution is positively skewed. ‘ Hs Kurtosis (B,) = = 2.9936 <3 B= OaF ‘The distributions platykxurtic. JS. | Find the M.GF. of the random variable X having the following probability density function x O 0 Distribution is positively skewed. (Que 36. | The following table represents the height of a batch of 100 students. Calculate skewness and kurtosis : Height (in cm) | 5961 |63| 65 |67|69]'71|78|75 No. of students | 0| 2 | 6|20|40/20] 8| 2| 2 Answer. Height) No.of |u=%—9"| fu | fu? fue pe | 2 (cm) | student f | 59 0 -4 0 0 0 0 61 2 -3 -6 18 - 54 162 63 6 -2 -2 24 | - 48 96 65 20 a1 - 20 20 - 20 20 67 40 0 oO 0 0 0 69 20 1 20 20 20 20 71 8 2 16 32 64 128 73 2 3 6 18 54 162 15 2 4 8 32 128 512 IN = Bf = 100] Sfu = 12 | Sfu? =164|5fu3 = 144]5fu* = 11! 3-8 C (CC-Sem- RAS eer ao Statistical Techniques Moments about 67; Ma = ny'= ny? 6.56 ~(.24)2 6.5024 + Bayt = 176-4 (11.52) (0.24) + 6(6.56) (0.24) = 167.19798 Coefficient of skewness, = (6.824448) 9.1694 (positive) (6.5024) Hence, the curve is positively skewed. Me 2 3.9544 >3 u Hence the distribution is leptokurtic. ahd at Pino Sight Apaaeldens A iespaantlal DPW Coefficient of kurtosis, f 3-9 C (CC-Sem.3 & 4 ee Qu Use the method of least squares to obtain the normal equations and fit the curve for y = “2+ ¢,vx to the following table x Mathematics - IV of values: x | O1 | 02 [04 | 05 een | || ee ¥ Answer Normal equations to the curve y =“ +¢,vx are x x 1 1 Le oly “LE 1 Dove = aLygtals Table of values is : 1 1 = fe a y ah ave a es o1 | 21 210 6.64078 3.16228 100 o2 | 11 55 4.91935 2.23607 25 o4 [7 175 4.42719 1.58114 6.25 05 | 6 12 4.24264 1.41421 4 1 5 5 5 1 1 2 6 3 8.48528 0.70711 0.25 1 lea 4.2 E(y/x) = 302.5} Eyvx = 33.71524 Eee Substituting the values in normal equations, we get 302.5 = 136.509 + 10.10081 ¢, BTD) (B72) and 33.71524 = 10.10081 ¢, +4.2¢, Solving eq, (3.7.1) and eq. (3.7.2), we Bet p= 1.97327 and ¢, = 3.28182 .e the required equation of curve is 1.97327 | 3 o8182Vz ¥ Henc ares, derive the normal WESTED] Using method of least sau equation to fit a parabola y =4 + bx + cx? from the following data‘ (hnaniinonintenieene 10 C (CC-Ser $-10€ ( *m-3 & 4) Statistical Techniques I Ly iu 17 | 20 | a | 29 Equation of parabola, y = a + bx + cx? Normal equations for the parabola are : a + BEX + cx? aSx + bix? + cEx? Exty = aEx? + bEx) + chet FS ay 2 = ay | 2 | Mt 28 4 8 16 56 (3) ou 51 9 21 81 153 | 4 | 20 80 16 64 256 | 320 | 5 | 2 120 25 125 625 | 600 | 6 | 29 | 174 36 216 1296 | 1044 | sx | fy Sky Ex? Ext ext | Ey | =20 | =104 | =453 =90 =440 | =2274 | =2173 Substituting the values in normal equations, we have 104 = 5a +20b + 90 (sn =5) 453 = 20a + 90b + 440c 2173 = 90a + 440b + 2274e On solving, we get = 11,b = 0.843 and c = 0.357 So the required equation of parabola is given by y = 11+ 0.843x + 0.35727 Using the least square method, fit a second degree polynomial from the following data: 7 Lelitl2isi4+5 +8 7 | 8 7 | 120 10.5 | 10.0 | 80 70 | 80 | 7.5 | 85 | 90 Also, estimate y at x = 6.5. jal,y sax? +x +e Let a second degree polynomial, y = * ; 7 : ‘The normal ications = the given polynomial are given a5 follows ¢ ache +b +e! 3116 (CCSem3 g 4 ne+ Sx +abx7 adr’ + bEx? + cEy nes at _3y. 0 0 0 1 1 10.5 8 16 20.0 27 81 24.0 64 256 28.0 125 625 40.0 216 1296 45.0 343 2401 59.5 416.5 512 4096 876 204 | 5x3=1296 |Ex4=8772 9 uations, we have Putting value in normal eq 204a + 36b + 9c = 80.5 1296 + 204b + 36¢ = 299 87720 + 1296b + 204c = 1697 On solving eq. (3.9.1), eq. (3.9.2) and eq, (8.9.3), a=0.18, 6=-185, ¢=1218 Then, y = 0.18x?— 1.85 + 12.18 At x= 65 y= 7.96 Que 3.10. | Fit the curve: po'=Kto the following data: [2 eg7em*) os) 1 [15] 2/25] 3 | vp ditres) _ | 1620 | 1000 | 750_| 650 | 520 460 Answer 1 j prs K Ae (Ey =K" p™ P t 1 1 Taking log, dog v = fee KS log which is the form Y=A+BX i where Yelogu, X=logp, A= lek and Bo“ 7 oo - oso —o9ee1e | 9.09062 3 0 o 0.17609 2.87506 0.50627 | 0.08103 osor0a 2.79239 0.84059 0.09062 39794 2.716 Toaoao | 0.15836 0.47712 2.66276 1.27046 | 0.22764 Ex = 1.05115]2¥ ere, . sul ee ituting the values in normal equations, we get 17,25573 = GA + 1.05115 B 2.73196 = 1.05115 A + 0.598825 B and ‘On solving, we get A= 2.99911 and B=-0.70298 1 1 yE-F = 1.4225: B 0.70298 oe log k= yA = 4.26629 = antilog (4.26629) = 18462.48 Asai, Hence required curve is pu! 42252 = 18462.48, Que 8.1. | Determine the least square ap} ax+bx+ctothe function 2*, at points x; = 0,1, 2,3, 4. proximation of the type Here saxt+br+e Normal equations for the given curve are, Byx? = alr! + bEx? + ee Eyx yy (Here m =5) ‘Table of values is, z 2 2 eos. a 0 1 0 0 0 0 i 2 2 2 1 m1 2 4 8 16 8 16 3 8 4 2 27 81 256 64 256 4 w | 64 Be = 10| ty =31 xy = 98 $x? = 30 | Byx? = 346 Ex} = 100 | Ex* = 354 Substituting the values in normal equations, we get 346 = 354a + 100b + 30 98 = 100a + 300 + 10¢ ON Mathematics - IV 3-13 C (CC-Sem-3 4) 31 = 30a + 10b + Be On solving, a= 1.143, 6 =- 0.971 and c= 1.286 y = 1.148x2 — 0.971x + 1.286 ‘Que 3.12. | Determine the normal equations if the curve y= ax + 5,2 is fitted to the data (x, y,), i= 1, 2, ..» m. Hence fit this curve to the data: y | 18] 51 | 89 | 141/198 Answer Normal equations are, = akx? + 3x3 and aS) + bExt ‘Table of values is z= fly] #2 eB Ei ay Figs! | 1 1g[ 1 1 1 18 18 2 bil 4 8 16 10.2 204 | 3 | a9] 9 27 81 26.7 80.1 4 | 14i[ 16 64 256 56.4 225.6 | 5 | 198] 25 125 625 99 495 ‘Total Xx? = 55 | Ex? = 295 | Ext = 979] Dey = 194.1 | Bey = 822.9 Substituting the values in eq. (3.12.1) and eq. (3.12.2), we get 194.1 = 55+ 2256 and .9 = 225.4 +9796 = 152 and) = Pe = 049 Hence required parabolic curve is y = 1.52x + 0.49 x2 On solving, we get Using the method of least square fit a curve of the form y= ab* to the following data : 2 oe sa|) 8) | Paes: |e » | 88 | 154] 331] @52| 1074 y=ab*® logy = loga +x log b ¥= logy, A =log.a, B= log,b $-14C (CC-Som.3 & 4) Y=A+Br Normal equations are, DV =nAvey & 6 M mt x Statistical Techniques I & 6 x, Y,= e SVAN a +B Ds teen =5) = y Y= logy ¥ 2 mY 2 8.3 2.1163 4 4.2326 3 15.4 2.7344 9 8.2032 4 33.1 3.4995 16 13.998 5 65.2 4.1775 25 20.8875 6 127.4 4.8473 36 29.0838 a A DY =17.3750 | 3x" =90 5 Lx¥, = 76.4061] & ‘Substituting the values in normal equations, we get 17.375 = 5A +20B 76.4051 = 20A +90B On solving, we get B= 0.69051 and A= 0.71296 0.71296 = log, a > a= 2071296 = 2.04 and 0.69051 = log, > b = ¢0-69051 = 1.99 es yeab¥ => y=2.04(1.99) '| Fit a second degree parabola to the following data : = 1 0 | 15 | 20 | 25 | 30 35 40 y | 1a | as | 16 | 20 | 2 7 34 41 We shift the origin to (2. variable x to X, by the relation Let the parabola of X= 2-6. fit bey =a + bX + By = am + DEX +cEX? EXy = abX + BEX? + BX EXty = aEX? + BUX" + cEX* 5, 0) and take 0.5 as the new unit. This changes the X?, Normal equations are : 3-15 C (CC-Sem, Mathemates-IV____—_———____““Sty » [3 x os Xy xt rr . 1 -33 9 99 | -27 Hi . 33 13 26 4 5.2 <8 i 2a ay 1.6 -16 1 1.6 -1 i 25 0 | 20 0.0 0 0.0 0 | 3.0 icles 87 21 1 2.7 1 || 35 2 | 34 68 4 136 8 A 40 at 123 9 36.9 27 ® Total = | 2X=0 [By = 16.2 JPXy = 143 EX? = 25fEN7y = 69.9] 2X9 = 0 ae Substituting the values in normal equations, we get Ta + 28c = 16.2; 28b = 14.3; 28a + 196c = 69.9 On solving, we get a= 207, 6=0.511, ¢=0.061 y = 2.07 + 0.511X + 0.061 X? Replacing X by 2x ~ 5 in the above equation, we get = 2.07 + 0.511 (2x — 5) + 0.061 (2x ~ 5)? which simplifiesto y= 1.04-0.198x + 0.244... ‘This is the required parabola of best fit. WHERE] calculate the rank coeffi ie a cient from the sales and expenses of 10 firms as given below : F Bal 7 les X 45 | 6 |50]5]45 | 0 [50 | oo | a0 [36 Expenves¥ [40 | 36 [90 |a]90 | a2 bas | a2 1201, }UeMonne & A) 4 hsathoblicnd Textaniegaan £ grace Repeated Rank of X column : 45 = 2times 56 = 2 time: Repeated Rank of Y column 36 = 3 time: Rank correlation coefficient, 3 1 1, ==(m} - + (m} - (om ~ m,)+ 35 ~ mg) +93 ~ a) ge NN?-1) éfos~ st (2° -2+2 -2+3' -3) ai 12 J 10/89) = 1 - 918643) _,_ 284 “~~ 990 990 r= 0.7636 Qusd6.] Calculate the coefficient of correlation between the following ages of husband («) and wife () by taking 30 and 28 as assumed mean in case of x and y respectively : 28 | 29 | 30 | 32 | 33 | 35 28 | 30 | 27 | 30 | 32 x | 24/27) 28 35 | 40 y | 18 | 20 | 22 | 25 | 22| 28 ioe] Given: x =30 and y= 28 g-17C (CC-Sem-3 & 4 ‘Mathemat z17 | 24) 18 27| 20 28 | 22 28 | 25 29 | 22 30 | 28 32 | 28 33 | 30 35 | 27 F 1 =5 2 25 4 10 100 16 40 Te ae | By 7P Be-2)G-7)) 217 = 270 = 159 35 | 30 40| 32] 10 4 Now we know that the coefficient of correlation is given as, pe ED 159 9” Tyg 3F xdly—y | V217*270 _ 159 © 242.0537 yy = 0.657 Que 3.17. | Ten students got the following percentage of marks in principles of Economics and Statistics : Roll Nos. 1]2]3[41516|7]8] 9 [10] Marks in Economics | 78 | 36 | 98 | 25 | 75 | 82 | 90 | 62 | 65 | 39/ Marka in Statistics | 84 | 51 | 91 | 60 | 68 | 62 | 86 | 58 | 53 | 47) Calculate the coefficient of correlation. in the two subjects be denoted by x and y respectively. Cia Statistical Techniques | rosy 1600 100 ~(ee" 10 x 2704) ~ (0s 0) = =0, VEL x 2224) ~ (0)? Hence, i PART-4 Regression Analysis : Regression Lines of y on x and x on ys Regression Coefficients, Properties of Regressions : Coefficient and Non-Linéar Regression. ee Questions-Answers “Lang Answer Type and Medium Answer ‘Type Questions between the two regression Que a8 | if the 0 is the acute angle show that tan = lines in the case of two variables * and y, Ke os. ' ngs. Explai Pr Ses where r 5,19, have their usual meanings. Explain +5, the significance of the formula ‘Answ r = Coefficient of correlation is given by; when r=Oandr=#1- | 3-19 6 (CO-Sems g » Oc n¥ dedy- 2 dx 5 dy re eens - (dy? where, dx=x-% dy =y—- 7 Coefficients of regression are given by, _ n¥ dxdy-E dx E dy 9 Edy? - (Edy? _ n&dxdy-E dx 3 dy 9 WE de® - (2 dx)? Equations to the lines of regression of y on x and x ony are 92 72 e-%) and x-F= 2-9) ox Sy ‘Their slopes are m, Since r? <1 and o,,<, are positive. ‘4ve sign gives the acute angle between the lines. Hence, Whenr=0,0= a “=: The two lines of regression are perpendicular to each other. Hence the estimated value of y is the same for all values of x and vice-versa. When r = +1, tan @ so that 6 = Oor 7. The lines of regression will coincide and there is a perfect correlation betwee? the two variables x and y. Que | In a partially destroyed laboratory record of ap 4 analysis of correlation. data, the following results are legible variance of x = 9. Regression equations are 8x - 10y = - 66 40x - 18y = 214 3-200 (CC-Sem.3 & 4) Find 1. Mean val Tl. Standar MI. Correlat lues of x and y ‘d deviation of y tion coefficient between x & y, Answer Since both the lines of regression bass through the point (Z, 9) therefore, we have 8¥- 1054 66-9 402 ~189 214 = 9 (3.19.2) Multiplying eq. (3.19.1) by 5, we get 40% ~ 505 +330 = 0 (BAG) (3.19.3) Subtracting eq. (3.19.3) from eq. (3.19.2), we get 329-544 = 0. 5 247 From eq. (3.19.1), 82-170 +66 = 9 Hence Now, variance of x =o? =9 6,=3 The equations of lines of regression can be written as Y= 0.8r +6.6 and x = 0.45y 45.35 | ‘The regression coefficient of y on x is st 08 (3.19.4) ro, The regression coefficient of x ony is oa Multiplying eq. (3.19.4) and eq. (3.19.5), me 70.8 x 0.45 r=06 From eq, (3.19.4), we get standard deviation of y, 45, (8.19.5) we get Que 3.20. | Find the coefficient of correlation (r) and obtain the equation to the lines of regression for the following data: =[6[2|i0[4]@ y|9|i1[ 5 [8|7 0 1 Regression coefficient of y on x, b= Regression coefficient of x on y, be EY 8. 9 SY? 20 Equation of regression line (y on x) is yo ow -26 y-8= pe) or, y B= -0.65 (x-6) or, y =-0.65r + 11.9 Regression equation (x on y) is x-z2 By 5 BY? x-6=-13(y-8) x=~13y +164 r= 0.65 x (—1,3) = 0.845, r= ~J0845 (As ) r=-0.919 both d,..,b,, are negativ” Correlation coefficient, ———___ Statintic Fal, | If for two random variables MF . x and y with same mean, wo regression th "earey=ax +h and x= uy +f, then show that he find the common mean, Here, [ot the common mean be m, then regression lines are y-m=a(x—m) by = aby =a y=ax+mQ~a) (B21) od x-m=a(y-m) x= ay+m(1~a) \ (8.21.2) Comparing eq. (3.21.1) and eq. (3.21.2) with the given equations, b=m(1-a),h=m(-a) following data were obtained Ex = 130, Dy = 220, Dx? = 2288 Ly? = 5506 and Dxy = 3467 Obtain the two lines of regression. : Regression coefficient of y on x, b= Ex? ~ 3467 L152 b> 9988 Mathematics - IV 3-23 C (CC-Sem.3 g a 3467 <2" = 0.63 5506 y—22 = 1.52(x - 18) 52x + 2.24 .63 (y - 22) x = 0.63y —0.86 ©OO + tical Techniques | Q Marks Questions) 3.1, What is the meaning Ane 32. of skewness ? ‘means lack of symmetry i.e., when The term skewness et ahd showed drt sand this distribution may be positively skewed or negatively skewed What do you understand by measure of kurtosis ? Discuss in brief. Measure of kurtosis Hw L(e-z) _ 2e-z) Bo Be me ON If, =3, the curve is normal or mesokurtic. IfB, > 3, the curve is peaked or Jeptokurtic. . IfB, <3, the curve is flat topped or platykurtic. Discuss in brief the types of correlation. ‘Types of correlation : Positive correlation : Ifa decrease in the value of one variableX results in a corresponding decrease in value of other variable Yon an average, the correlation is said to be positive. Negative correlation : If the decrease in the values of one ‘variable X results in the increase to a corresponding values of Y, the correlation between X and Y is said to be negative. . Linear correlation : When all the plotted point lies approximately ona straight line, then the correlation is said to be linear correlation. Perfect correlation : If the deviation of one variable X is proportional to the deviation in other variable Y, then the correlation is said to be perfect correlation. 5 Define regression lines. Aline of regression is the straight line which gi fitin the least ojuare tothe given feaueney net eves oe correlation, ALS, Coefficient of correlation, 3.6. Write the normal e Square method, uations to fit a curvey = ax? +b by least x ane yee Y=6X+a, wherex= + andy=y x Normal equations are LY = na + bX EXY = abX + UX? 38. The regression equations calculated from a given set of observation for two random variables are x=-04y +64 andy =-0.6r+4.6, Calculate mean values of x and y. ‘AWK Since, both lines of regression pass through (Z, 9) So, 7 +0.47 - 6.4 =0 065-7 +4.6=0 Solving by cross multiplication, z (0.4 x 4.6) —(6.4 x1) a SQ-10€ (CC-Se &4) Statistical Technigu 39. The first three central moments of a distribution are 9, 9 5 0.7. Find the value of the moment coefficient of skewnes,” £25,201 foment coefficient of skewness. Given: », = 0, To Find We know that, moment coefficient of skewness = 0.177 Differentiate between skewness & kurtosis. Skewness Y - Kurtosis 1. | It describes the tendency of a| Kurtosis means the measure distribution that determines its | of the respective sharpness symmetry about the mean. of the curve in the frequency distribution. 2. | Skewness measures the degree of | It measures the degree of lopsidedness in the distribution. | tailednessin the distribution. 3.11. Fit the equation of straight line from the following data: a ae |e ois | a |e |G y 4 | 27 | 40] 5 | 68 wus, Let the straight line of best fit be yratbr (3.11.1) Normal equations are (3.112) and 113) Here m=5 Table is shown below : Se 1 2 3 4 5 [x= 16 ‘ys matics ~ IV (2 Marks Questions) $Q-11 C (CC-Sem-3 & 4) gubstituting the values in eq. (3.11.2) and eq, (3.11.3), we get 204 = 5a + 15b +(3.11.4) 748 = 15a + 55b (8.11.5) Solving eq. (3.11.4) and eq. (3.11.5), we get a=0,b=13.6 Hence, required straight line is y = 18.6x ©OO

You might also like