6.a) What Is Interpolation? Prove that (WBUT 2013(0DD)) Sk)= yy +tay, + ee Ay, Mee D ary swat) Ay, Answer: Interpolation: In the -field of Numerical Analysis, interpolation is a method of constructing new data points within the range of a discrete set of known data points, hi In some cases, it tries to approximate a function from a set of given data points. This is known as curve fitting or regression analysis. Let us consider yo.yi,.-.Ye be the values of a function f{x) with respect to the values of x Within equal intervals (XopX1y...Xa) Where X-Xet=h. U8) = 64 + 6)(3~ ag) +O — Bo ME AH) Feel XA ANE Hoel Em Spa) Whete x= xo, = flxe) and La(t0) = co 'NM-33 POPULAR PUBLICATIONS From the above equation, . Putting xx in the above equations and considering that La(x.)=Mx.), we get, )_ Af(n) u 6, = LG) = 1% (0) _ ATL) niin ain = 1 SEO) = fad b I-%, h Yl) pea fx) Lorem rem ot ae oth EA Z2yA" (40 No I WL) Neat git Now, u=222 and usr Then, m Or equivalently, vy tuAy,, [ME ante — 1A yy / 286 + (u(u =I). n+ a"'y, Jat s called Newton-Gregory Forward Interpolation Formula, Thi b) Prove that V'y, = V" Jor [WBUT 2013(000)) Answer: We know that At = E'V! = VE", Therefore, Avy, = VE" (yx) =W'(E" yk) (proved) EV,4. Derive Lagrange Interpolation formula. [WBUT 2011(0DD), 2016(0DD)) Answer: a a functi = f(x) corresponding to the IE Axo), ‘Rs fxs) be the values of a function y = fx) correspon i oe ‘ yen are not necessarily equally spaced, then the Lagrange’s interpolation polynomial is cor) flr,) 1) where L(x) =} ————. seseee( 1) Whet x (xx, Jo") (x)= Tar, )y esseas( La) and ro O(x,) = (x, —xy (x, (x, Proof Let us form the fundamental Lagrangian function (X= Ag X=) Joosel¥— NTA Ips (x, — 9x, - 4 @, (x)= a(x) » (r= 5)0,) ‘The Lagrangian function «o,(x) has the properties that w(x) => a sesa(2b) msl, ifrss Let, £,(x)= ¥@,(4)f(x,) whieh is a polynomial of degree n’such that , LG) = Yo) =f) satisfying the condition (/) of interpolation. ‘Therefore, L(x) is an interpolation polynomial which is unique. 2. Defina internolation and extranolation. Deduce the Newton's forwardFO)* fO)+(8— Hf 0008) HEF So) HORE HE) rtp) 34 (= 2) 204 (= 204) x45 4x2 4Ne—T) x1 =34+20x—40+45(x° —6x+8)+13(x? ~ 13x" + 50x56) = 13x" -124x7 + 400x405 21/176 18. Prove the following operator relation: I, #° =1+75*, where the notations have their usual meanings. [WBUT 2018(0DD)) NM-20, NUMERICAL METHODS Answer? We know, 44 S=Et-E? and u Now, S7=E+E'-2 Vogt of m =F +2 (l) and og (Ee +2) | ay" +242) fusing (1)] = gr ea)ated s 49, Evaluate the 6 missing terms In the following table: {WBUT 2015(0DD)} 0 41 3 4 5 14): 0 - 3 15 35 Answer: la f(I)= and f(4)=b Now we construct the difference table: x T(x) Af (x) A(x) Af (x)45, Evaluate A? tan 2x. [WBUT 2014(0DD)} Answer: AM tanx=(E-1) tanx =(E° 26 +|)lanx =tan(x+2h)—2tan(x+h)+ tans, where is the step length. I (21) ate 16. a"|—|= . BUT, 2014(0DD) ¢) x(x h)(x+ 2h x+nils) eBur 20r4fo00}) Answer: aft}t tech mh Ih x +h) eh ox x(eei) xreh) xx+h) So, the statement és true form =1 ‘Ow we consider it is true form =k 1) (ayia at ()- x) XE WY ool NM-19 LIC: iS alt (4) = ofe (4) =At (4) at (:) x x. xth x, : (20a! (cnt ktit © (x AY et WYK 3h) et K+ DA X(X+ AMX + ZN)oorox+ kh) : (yt kant f too - nde ti) Lee (Ih x xoxn(kt | (x kh) | x(x+(k 4 Dh) a Cyt piat! XRAY EH Uh) next RAYS + CK + ID) «The statement is true form =(k +1). . Hence by the principle of mrthematical induction we-can conclude that the statement is true for all values of m.From the definition £, (x) = where wa)=T]r-x), D.=(s-x)wits) * ne ; ina) = (1-0)01-29(1-3)(1- 5) = 8 a0 [a2 "3 = ‘ ° mes DT =F) | £6) 2 -30 0 as =30 ~« |8 a. =6 in Is 5 2 -l20 | 35 3. =120 NM-16 NUMERICAL MErHOns fide ae sees #08 = 1) =-Bx(-0.375) =3, 4D. tAV=A-V. aa ™ INBUT 2014(EVEN),2014(00D),2018(EVEN}] apse Aye) Seth £0) & (x)= f(s)-f(r-h, teres the step length, therefore, ae VE) =) FC) . = f(eth)-S()- fQx)+fle-h) = f(e+h)-2" fa)+ f(x-h) gssso)=S/@)- Sh) eAfix)-Af(2-h) = f(xth)- Se) - £0) + feb) 2 fist h)-29 f(x) Se~M) ., From above two, (B-VPO)= A WG) 1b-VeAV 44. The function y =sin x Is tabulated as given botow: [WBUT 2014(EVEN)] : 0 ge. & ne 4 2 sins 0.70711 1.0 Find the value orsin( 2) using Newton Backward Interpolation. Answers yesins 0 a/4 1/2 y 0 0.70711 1.0Pe stomata) (> 9 a 2012(ODD), 2015{0DD), 2016{EVEN), 2017(0DD)) ots e(1+4)(1-¥)£(6) =(4a)[F)-W(9)] =(1+A)[70)- £0) +7(8-4)] =Ef(x-h), vltAsE =S(2) =b-f(x) Hence, (1 +A)(I-V)=1 in . & Show that if A operates on, then af, "}-(3). and hence , i". mel) (1 [WBUT 2013(0DD)] sla) (ee et] . Answers : Part; Liss =" = 78) _( 7) 20'.="Can st) (eat) (et, (oe mt n! fat 1 “[etlb=s [eelncact leellusaciln-» : NM-15 Popul jONS. mt fat ie(n-x)]__ ot atl *Qelmoxal| nex | beelinoz—t wt ine {"e “Wax <-(") R.HS (proved) 2" Part: x" )- Sal . ) {using the previous result] at, xtl easma( "a(t ea(ialemtlen) ("3 i SHCA} Cams ' (.: -(e4) (oa) Bf) os1.222x! = 0,889.27 +3.278x" -3.444x + 2.933 1 | show that the divided difference of ni” order [WBUT 2011(ODD), 2015(EVEN)] 5. Given the funetion y=" pense TCL) erates) Answer: : tl Bos oe) oy MRI Te Ril) Ba I =I) Similarly, yx, i Then, Hered =F Sar That is the result holds for =I, 2. Let the result hold for a particular value mof n. Then, ¥(x99% 9%; Now, ¥(Xo0% X20 | So the result is true for n=m +1. Hence the result holds for all values of 7, that is NY. (Proved) i VK ogo) = 6. Fit a polynomial to the following table of values using Lagrange Interpolation formula: : (WBUT 2011(ODD), 2015(EVEN)] a" x: oO 1 y: [a2 fo |e | Find the value of y when; a) x=2; b) w=3.5 . NM-14 r : NUMERICAL METHODS asst ig 10 Lagrange’s interpolation formula the required interpolating polynomial is” Neco DEH DEA (1) EOE EA), 1,097 @= 10-304) (1-0)(1 31-4) (x= 0)(x=Ir-4) (x= 08-1 -3) Goo cna eae 2 (8-8 NN— 4) = (ONG = (= 4) + (= ONG N03) 2 (e108 30-4) -xx-D[x—4-2 43] 2(e-D[(r-30-4) +3] =(x—Ix? - 7x4 1242) nox 4 12v—x7 + 2K-12 Hx? Tx? +19x-12 \ es amd 10/9519 kt.ayProve that (4) = /(3)+/(2)+47/(1)+427(I)- DBUT 2008, 2044(EVEN) Hoc hve to poet {4)=/0+F QS 7/0410, sere f(x) = fe +1)~ fla) and Iris the step length. eB {M=8{Q)-V LO Safer f= 82) Safa’ f)'sf)=¥O)=SO-JO) IG) MDO LIES L0= LA 2 fA)= JO)FSOD+8 SO +8'70. Apa) Af) by Evaluate a" e0s2r- [WBUT 2008, 2014(EVEN)) ‘Answer: ¥cos2s cos 28) = A(eos(x+h) 052) alone uth) 2 =A(2sin (r+ Apsin(-H) =-2sinh A sin (2x+A) ==2sini(sin Qe) 8)~sin(2e-+N)) sin: 2.) What Is the difference botween interpolation and extrapolation? Give suitable examples. IWBUT 2008, 2008, 2015(000}] What erences between Interpolation and extrapolation? Explain with suitable examples, WUT 2017(000)] Answer: Interpolation is the process of obtaining a value from a graph or table that between major points given, or bet used to obtain the vale, Extrapolation is the process of obtaining. value from a chart or graph that extends beyond the given data. The "rend" ofthe data is extended past the last point given and an estimate made ofthe valve. Example of Interpolation: dra hhave to the last allows you to « between any data points that you have) i. filling NM-10 located datapoints plotted. A ratio process is usually ‘a line (fiting & curve) from the first data point you ate dala points belween those two extremes (or in between". NUMERICAL METHODS ante of Pxtrapoaton: fing a curve othe dt ’ ’ iholt then) beyond the measured dat, ——— extptation OEus When you ae neg ore mi celle iis (AeteD) bat we oui Tis occas When the ealeutated Cones 7m ion grin te ealeulted Concentaion > Maximum Sando Conenteson for total biasay standard curves, extrpolating cat be very dangerous and. que often Be ileaing, The resson being that minute changes in MFI values onthe fa pas ofthe ‘indand cirve can lead to huge changes in concentetion or ove inerpoation occurs when your MFUIRLU values are within standard range or Minimum sandard Concentration < x < Maximum Standard Concentration. [daly this is the Sys where you would want all your unkenovn points tee fing concentrations for points that le of the range of our standard MiPenmdictenve Tennant 8 CConcentration/Dose Extepotation vs, tuterpotatton The ony difference between interpolation and extrapolation is that frst calculation is” made for arguments inside of arca covered with experimental data and extrapolation for ‘outside range of tested values and meant to be used for prediction,