You are on page 1of 20
For example, suppose X is a mixture of an exponential distribution with mean 100 and weight 60% and an exponential distribution with mean 200 and weight 40%. Then the probability that X < 100 is @= 200 160/00) poexcio Flos) =0.6(_e dice 450 Cire )=0.5367 For example, let X be the cost of a dental — claim. This is a single claim and X has a distribution function. However, this claim could be for preventative work ~ (cleaning and scaling), basic services (fillings), or major services (crowns). Each type of work has a distribution Xi). If 40% of the claims are for preventative work, 35% for basic services, and 25% for major services, then ~~ the distribution of X will be a weighted average of the distributions of the 3 X;'s: Rec}=tatber}H+ Goalie) + FQ): 0.4 FG JrossFbu)40.2sF&;) Comp ond Howeney he Uariana. famixtrre is nok fhe wesiaied average oF Mo vodantes Yeu. i Second homerk onl He cokvaslina fle sajtens of Hua, aan — ExampLe 4A Losses on an auto liability coverage follow a distribution that is a mixture of two Paretos. Each— distributior ~~ has param ne distribution has parameters a =3 and @ = 1000, and the other 2S anho, Ss Zz >in the mixture has equal weight, eters a=3 and @ = 10, —— Calculate the variance of a loss. = bok X vop = [208 Rao EU =05(att) tes Get and EX Is 05 CEE) vas (Et) 0 5 (ED), 0 5 Jor lows S20 {RTE Nor CX) Usher Xv Pareko CX, ror=3 , @=100 2022 ) sos G2 tes (S2")-2450 Lgenn?Ca)) 7 = So So00e0 z i 50,800,000- (24s0)"= 42,437, 500 ~ EXAMPLE 4B The severity distribution for losses on an auto collision coverage is as follows: RIE: ECX = pe eee, = Conny ( 7500 ) : «\z000-+x) ~Wik\ 7500-+x 35 OQ, =207° A=) O,=FSe0 Calculate the coetiicient of vatiation of loss size. = N= GH. ld X yep Lass sizes: x> Pareto 30 0.2 BP) ~ 190 EDO] bs CP). 03 GESE,) =s,u2 5,000 So Ver Ch) = #425000 ~Ciys0)*= 6222, 520 1 Cve JG322, 560) SEAR SOO sO = L434] ExampLe 4G On an auto collision coverage, there are two classes of policyholders, A and B. 70% of drivers are in class A and 30% in class B. The means and variances of losses for the drivers are: ‘Class [ Mean | Variance = A | 300 | 30,000 — B_ | 800 | 50,000 Aclaim is submitted by a randomly selected driver. Calculate the variance of the size of the claim. _ PS RTF Now Ot) whee % Kup dain size Recah Vor6d) - EDC] - EDT EUS G4 (ce) 0.3 Gos) y50 EDC] = Vo-Od+ ECx* ECX*] =| 0.4 Gogo + zon" )+0,3(s0p00k 8as*) =241 000 llnce Nev OO) = 241 900- yso" = 88,5 ~~ EXAMPLE 4D The number of losses on a homeowner's policy is binomially distributed with parameters m= 5 __ and q. The parameter q varies by policyholder uniformly between 0 and 0.4. = Calculate the probability of 2 or more losses for a policyholder. RIEL P CK 2) whore Krep bosses awd XrBin Lm =5q) Pex 42) ~1- CPCxa0) + POX) = G)q° Ling)? =(- 9 =G)q! C-g= sq Li-g)! tHiso b ony £ oy PO%ce) = Bra CPG) be~ ay 5 nay ay C =) zy L- oe a % © av “A O-b.¢) 4 ms ‘ i | Cinq)" deb q, soiz,s) Cut wda ° oe ~aas/ af aye) =2.sli-o J)- 36-28 So ecXbe)ep ilar )= flog) $2 s(-0,¢) ~ ¥ li-0.¢Ja2s (0.8) - 22 (ole) = 2.3056 Lssei=0.4162 hb pOXx2) <1-0,916¢-2, 28 Exampte 4E For a population following a frailty model, you are given @) a(x)=1 (ii) Ahas a gamma distribution with a= 0.2 and 6 =0.1. Fora randomly selected individual from the population: 1, Calculate the probability of surviving to 70. 2. Calculate mean survival time and the variance of survival time. ) Since A=) Yeon AG) = (dre =X and SG@ | so _ et o fulisStle MEE of Gamma? My(D(-o b) ss MM Lreid) = MCx) Hors C0162) **=[ =o, 2 =0 25 LY pe T Ye] tS pro, 1X, oto, c ~~ ExaMpLe 4C On an auto collision coverage, there are two classes of policyholders, A and B. 70% of drivers are in —— __ class A and 30% in class B. The means and variances of losses for the drivers are: Class | Mean | Variance A | 300 | 30,000 B_ |} 800 | 50,000 — Aclaimis submitted by a randomly selected driver. Calculate the variance of the size of the claim. o}- sep clodin st20. ond et indicate 2 Pre noms using Contilioncal Maniac Formula f Veratolb_| Near OO) Naw CE EWar So Vor bathe class urhp.e L~Ramroull: Vor CEL X}Z)) + ELVae(xIZ)) 122) = (6.2)G.3) on -200!* = 52, Iz) J=0.7-( a5 e00)+ 0,3(so¢n) ~ 36,000 X)= 52, Soo +36 jor = FB, sc Using Banserlhi Sroskack’ EXamPLe 4F Claim sizes range between 0 and 1500. The probability that a claim is no greater than 500 is 0.8, Claim sizes are uniformly distributed on (0,500) and on (500, 1500). Calculate the coefficient of variation of claim sizes. TE :|CV = AbM the daunsize and either IebxX> san o X45c0 KT=d if x € Co, sea] amd Tai if X€ (Senso | Since _X4 basi fore ECX( kes b= SSP eso and ELXIX> sede PE ce Given. \-9.% Le Ddi=bCe lxigys oe (ase) +0% Cian) = uo0 Nar C oF Ly bse) = SP 2 ate Lx] Kw Je ESR * Ver WIL) 0.20 “te )to 2 Ce >) was, be Var (BOD) uscngy Burnout. Shokeut = (.8)(o2) Gove - ase) =40,000 Fi 1. Norrlox)= qo20 + 33,337.33 ayolCV= ~'23333-73 = 4 S480 Uae = [23,333 An Boch ple ushoe inclicatne voile) Tis Combieuiens —— Exampte 4G (Same situation as Example 4D.) The number of losses on a homeowner's policy is binomially dis-— __ tributed with parameters m =5 and Q. Q varies by policyholder uniformly between 0 and 0.4. / Calculate the variance of the number of losses for a randomly selected policyholder. ruslowy of Neve, thalndicade- vamalsly is Q RIEL Nar CX) cdbare Yee | EDX Ta\mQ = 5@ and Na OX) = mQC-Q)=sQ(1-Q) D Nal Ox Nes* Vorb 25 Ver Efe) 1100 (500+ x} a x<100 500_\* 100( re) x> 100 Calculate the mean of X. Mixdusps: Y.-B; Proddey Medahs 24.4 =ta jCoudibionl Vee: 4.12~ bil s Spices: Y-20- 4 2s Mixture You are given the following information about a portfolio of insurance risks: + There are three classes of risks: A, B, and C. are given in the following chart: Number _ Claim Frequency 1 of Standard | = Class | Risks Mean Deviation A 500 0.10 0.20 tI B 300 0.12 0.25 = c 2000.15 0.35, Determine the standard deviation of claim frequency for a risk randomly selected from the portfolio. (A) Less than 0.240 Total. thof vi S ns ks = soo: ~ (B) Atleast 0.240, but less than 0.244 af nsh ia — (©) Atleast 0.244, but less than 0.248 (D) Atleast 0,248, butless than 0.252 (E) Atleast 0.252 RTF stddev(s +) + The number of risks in each class, and the mean and standard deviation of claim frequency for each class, _ £60 ZOO Ec dk 8. (.)+ 45 (0.1) + aS Recall | E cx] 7) Ug 52 Co 22to,10 )+ 3 (0,25 b01Z)p = 0,07707 7 Sslddda = J Nowad =_) 0,0990%- 0.167 = [Orde 31 = ~~ 4,2-8, Use the following information for questions 4.2 and 4.3: z 2 e335 to fe) 1s [Fora group of 1000 insureds in three classes, you are give! | Numberof Mean Standard deviation _| policyholders _ loss of loss 500 10 12 | 300 20 30 _| 200 30 60 TS55 — 4.2. The number of claims submitted by each policyholder is identically distributed for all policyholders. — 1000 claims are submitted from this group. Using the normal approximation, calculate x such that there is a 95% probability that the sum of the claims —isless thanx. Mexhuve 4.3, Each policyholder submits one claim. Using the normal approximation, calculate x such that there is a 95% probability that the sum of the claims Is tess than x (Sur) 42) Xhep dais whe Xss ore tid ; RTE: P (lowe K <2) = 0-45 selemex a = W.(.95) 50 \cao XA Eocds eo (16) + “ee (20) + “wae Co)= 14 C1 wv) Se padi) = 00 (Ror tom Hus) = 2 a — 7 Boel oc (i G)+0.3 oy ot) +0. 2(Gp*+ 30%) = 1YI2 COs wie- ites er wh Cooo)= 123,000 far 1x Husé Hence| o- ~ Jn Coo Tuas (se X— goon = |.64s a Vv W JG 7. L.Gys (uae) + 1goun = 1893.23 463) EIXTE soo Co) } sen(20)+ 208 (30)= 2000S myer of Ye sur Nok ¢ uve: a bbicncp of he means Var Deassoli2 + 305 C20") + 250 CG8") = 1,062,0000 Venaneof fre = sun = Sur of He Genomes S RYE: PCX40¢) = 0,45 2) X14 = 06,45) 5 X- Ione = 1.GU5 = Jio6z,c0' fo = Reve +1.6U5 Sica2e00 ~ 18,695,235 (Nonanen s lower thon in dey OO ca has bed bedi. perncved) ~~ 4.4, You are given a portfolio of 100 risks in two classes, A and B, each having 50 risks. The losses of the risks — in class A have a mean of 10 and a standard deviation of 5. For the entire portfolio, the mean loss is 20 with a___ "standard deviation of 15. Calculate the standard deviation of losses for risks in class B. — (A) Less than 9 = (B) Atleast 9, but less than 13 (©) Atleast 13, but less than 17 a — (D) Atleast 17, butless than 21 — (Atleast 21 “This ts ie y bre loss = 2 Ec De20 = 4 Cio) eC.) =2 20 = Cor ua) Uo =10 Hirg 224g =3e Atso E08I= Var Or FOC = 1st#20° (Por ere potato ) = 625 so 2 (o%+s*) 2 ( ope )=625 22C + 5 ~~ 4.5. Losses for an insurance coverage follow a distribution which is a mixture of an exponential distribution — — with mean 10 with 75% weight and an exponential distribution with mean 100 with 25% weight. — _Caleulate the probability that « loss is greater than 50, LabXveg loses wlune X~ mixhue (Bug io 210 WW, =20FS, Expire (os) = tSO wo =05 eyes P& >s0)= CCK >a) OC K, > 50) Co.25) PCX, > ss) = SCsa) = J Cine | we PO sw 00g PCX, > $:)=SGo ee "P= © Go 531 2. PCX>se0)=0,75(9,00673¢)+0 25 (0,@068%)) = 0, 564 5 Ci-w) 4.6. Losses for an insurance coverage follow a distribution which is a mixture ofan exponential distribution — with mean 5land anexponential distribution with mean @) The mean loss size is 7.5. The variance of loss size is 75. ew ~~ Determine the coefficient skewness of the loss distribution. — Skerry, EC x )?) ~ wL : 3 < = 2 Oe =? om 2> pO = 8) Expr O= {°; ~?) Swen Hock moon loss = 3.5% bet Ls be dhe reign tofexp vol ven onen © bO-w)+ow = 25 5 5-swtew 2.5 2 wle-s)=25 Nox as @ Givers that toe sonenrceol lov sien = 35 Hho a TE TX7] for ath losses = Vor OO4 EL *= Fo +R Ss = 13,25 Chron Tits) EDC: 207 anf ELX?3: 2 (s)t 2 50 so 50 (ilo) au ot ~131-25 =) So -Sow time = (31.25 = sou P2wWEr= KLIS ay 2w Ct - ase ¥t25 sub) Into © 22(3) Cors)le-s) e125 ~ 5 (ors)=en 25 Sotadetias 4 so seas 02 125 |Plg me © ~w= Soe3 oy SMunmesh ECD 36 ab 2yd hue EDO3= 60? so LOOT: Loc 6) boul y.23) - = 3R6R EIS Skungay = 3864.1FH5-3C131.25 Gs) +2045) @sy" 2240633 - 4.7. Fora liability coverage, you are given * Losses for each insured follow an exponential distribution with mean y. + y varies by insured. + follows a single-parameter Pareto distribution with parameters a= 1 and 6 = 1000. Calculate the probability that a loss will be less than 500. Roe. POX 6500) Uline Xreplsrer and W~ Exp Crs: a1 ee) CPrewtolbtes) PCX ¢sen)= a J Foe [CGS dye LOS] J Tv 4 100 “ ine Ta Io oog ne Ty lo ‘poo pes =F pach cs/ seo — =e this clousity S-24+ Leese tle =r-lb2e * +0213) "4.8. [151-82-93:11] (2 points) A population is equally divided into two classes of drivers. The number of acci- __ dents per individual driver is Poisson for all drivers. Fora driver selected at random from Class 1, the expected number of accidents is uniformly distributed over (0.2, 1.0). — _ Fora driver selected at random from Class 2, the expected number of accidents is uniformly distributed over (0.4,2.0). Fora driver selected at random from this population, determine the probability of zero accidents. a) 0.41 | (B) 0.42 (C) 0.43 (D) 0.44 (E) 0.45 hak Yeats numdaer of octiclenks uh XU CX) and dys thai (0.2, ) RTE FOS) ples Conta?) har X]20 5 paf of Risser =o) ee ee rs} Now PLX a= (eta el vaste [27 2 |rasarG ~ tate few e t L 5 a [Lo 2 the cupis Cosa Gt osu) = 2.449 (6) oa ~ 4,9, Ina frailty model, a(x) =2x and A has an exponential distribution with mean 0.5, ~~ Calculate E [X?|A=0.49]. a a Dee Mod =x* and SGI Nzo.ua)- ee PO perbles o Wo bullu/ O=aeF avcl P22 '. EU *1(s50 (42) = tga (On) =da9 1) = = GT rf ot x? ai] = 349 feb bas drag (=te-* os etatl 8 “ 0d vt Foret) = ee de Jo. assoc oO ~~ 4,10. Ina frailty model, a(x)= 1 and A follows an inverse Gaussian with = 2, 8 =1.— Calculate F(0.5). = AG As tues Guasion lu =2, 0-1) Recall SCs) ~1-F0,5) se Ces) = 1-SG.s) SCos)+M_ Coss explo [ ia) a f — {te re 2a At J = e s) | = explG G- STG)? Co.5))) =explo C- Js ae =°,534100 F(b.3) =l- 0.53900 =p. 6 100 4a Survival time X for a population of 100-year olds follows a Weibull distribution with the following hazard rate function: — h(x | A)= 3Ax28 — A varies over the population with a gamma distribution having parameters a=5, 0=0.5. Calculate the marginal expected future lifetime for the population. Xosetie* 4.12. The size of a loss has mean 3A and variance A?. J has the following density function: f(x)=0.00125 (= x > 4000. We Calculate the variance of the loss. Hess A sirqs Pore Parehs Z ~~ (A) Less than 22,000,000 w/O=YORD 50 =5 - (B) Atleast 22,000,000, but less than 27,000,000 ~ (C)_ Atleast 27,000,000, but less than 32,000,000 (DP) Atleast 32,000,000, but less than 37,000,000 CA!" Ve EI - — (BE) Atleast 37,000,000 _ Ack XV ize Loa | = 23> and of = Gta = oo)! = = sol 4 EC@I4 sCyom)* = yoesonoe S-2 = LuNJarf{a)= 2RSIEOO_ et. BOSSECe Ve OF Gai Nenana = ECL) ever C sty eOt]+ 2 Nlo-(3) = SoH Voead) CASI = 4 41 bible , 66 © __ 4.18, [4B-S95:14] (8 points) You are given the following: probability 1/4. 1/3. — _ 0r 160 with probability 1/2. Determine the variance of the pure premium! for this risk. Fora given risk, the number of claims for a single exposure period will be 1, with probability 3/4; or 2, with Ifonly one claim is incurred, the size of the claim will be 80, with probability 2/3; or 160, with probability — Iftwo claims are incurred, the size of each claim, independent of the other, will be 80, with probability 1/2; A) _Less than 3600 LN be Yee rear oF doumg Log — (B) Atleast 3600, but less than 4300 =) (©) Atleast 4300, but ess than 5000 STHYAS eres Per — (D)- Atleast 5000, but less than 5700 (E) Atleast 5700 hod PP be de Pure Rewinen > RIE Nour CPPI= Var CEC PRINT| + ELVar CPPIM] Using, bo Bowroulle Shoreuth: TEN EU ton ECPPINT = $(50)4 E(1c0)= 22 FE psp thon EcPPINW J =e) 4 $C ie) =120 C1 Chains) So 2 bcm ELPPINI =2C128) =2H0 Shobouk De Ner CE CPN) tlm) ¥ecw=2) + CELPPIN=1] — EPP/A =27) = CE) Gh) ( FB 3y0)*=2 33.3 v Ulan Me than Now CPP\ N=) = C4 OS) C eo -ge)*= SREP ele Loy =2 Yoon Ver LPP|N=2) = Ze CAR ACE) Ceo -80)4]~ 3200 eS LET Var CPIM] Jag CS) +t (S200) = (26.66 [ern ce. CPP) 3233334186664 -=S200) ~=4.14, [4B-F98:8) (2 points) You are given the following: * Aportfolio consists of 75 liability risks and 25 property risks. * The risks have identical claim count distributions. * Loss sizes for liability risks follow a Pareto distribution with parameters @ = 300 and a=4. * Loss sizes for property risks follow a Pareto distribution with parameters 6 = 1,000 and a=3. Determine the variance of the claim size distribution for this portfolio for a single claim. (A) _ Less than 150,000 ~ OA FP ae (B) Atleast 150,000, but less than 225,000 Xv Pareko G t art) (C) Atleast 225,000, but less than 300,000 x p~ Ruzto & = (00 04 ~ 3) (D) Atleast 300,000, but less than 375,000 (E) Atleast 375,000 lk L be Meindice ta Ss Sud wher Hk f Lradpi diy 2 Grerperby ovncl lok lac He hatin Size RTE. Vb OO) : Vew OC) = Var Le tarred], ED CXL) Shy beusol condbin Gave} ElAtltality l= @ = 30 -200 ~16> ‘ \ ECX*lLabilie l= @DOD =e ig) ae aA a OS Se 2 o "ar [het Linbilihy I= 72 - 100? = soe XN, 2 GOK Rogen | = SE peo and LUX" | Cope ty| Sts S Var LK Fogo = lows ~Son* =FS50.con = lope Mso Vn CETX ity 3) EGov - loo) = 39000 and B [War (KITT) = 2 (29, owt+i (4 50,01) = 202, soo “Thus Vb CX) = 39 00+ 202, S00 = 232,Se0 © 4.15, The number of claims on a policy has a Poisson distribution with mean P. P varies by policyholder. P is _ uniformly distributed on (1,2). = Calculate the variance of the number of claims. “wae (B) 19/12 (0 5/3 wm 74 B) 23/12 Ack) eh Hee nudes oF Caring police Nix Po LP) urkne P~Unibon C12] Rte: Un-ln) Vor We )= Var CECw (PI) + EG CrP) | telb-a) + + Cota) = = Var CM|PI= 3. \aCA)=Nar(P) + BOPI= 4,16. Claim size is exponentially distributed with mean A. A varies by insured, and follows a Pareto distribution — __with parameters a =5 and 9. Variance of claim size is 9.75. _ Determine 6. wa (B) 5 (Cc) 6 (D)7 ®©s hak K\d Ser Apis size rare Wr Ep Od) ond \~ Prctole=s,0) Loe ceive taka CX) =4.#57, RIF Now 0d) = NorCECXLY au(Xjs)]~ 44S 9.45) = Nor OM) EC] Srate fra Bees seo = cual Woirs S* Now BLY I=2 Gy ered = Or) _ zat neo Vox (= 23° - eI: 5 ae =q.9s= 2 + ( eS) -= eve* => @ tT 3e os 4,17. [4B-F92:23) (2 points) You are given the following: * A portfolio of risks consists of 2 classes, A and B. — * Foran individual risk in either class, the number of claims follows a Poisson distribution. — a Distribution of Claim Frequency — _ Number of Standard = Class Exposures Mean Deviation I A 500 0.050 0.227 — _ B 500 0.210 0.561 _ Total Portfolio 7,000 Determine the standard deviation of the claim frequency for the total portfolio. (A) Less than 0.390 ~~ (B)_ At least 0.390, but less than 0.410 _ —— (C) Atleast 0.410, but less than 0.430 — (D) Atleast 0.430, but less than 0.450 (E) Atleast 0.450 cls zt Fs Op Werte) Nix Po(= 0-650 \ar(10)= x eC Et Nace) (22 (Xo) 0210 -0 60) =(6 se so 1d) 2.0064 Eada] Db tes Co.229") Aero lo,561)*= 0.83125 cver(N) =o ISBIIS +0,0664 =0,\%q s Tus, = Doiaaeas' = 0.4352 Q) Alemalie Sok leg "4:18. {C3 Sample:10] An insurance company is negotiating to settle a liability claim. If a settlement is not reached, the claim will be decided in the courts 3 years from now. = You are given: + There is a 50% probability that the courts will require the insurance company to make a payment. The amount of the payment, if there is one, has a lognormal distribution with mean 10 and standard deviation 20, + Ineither case, if the claim is not settled now, the insurance company will have to pay 5 in legal expenses, which will be paid when the claim is decided, 3 years from now. ~~ * The most that the insurance company is willing to pay to settle the claim is the expected present value of — the claim and legal expenses plus 0.02 times the variance of the present value. 04. + Present values are calculated using i Calculate the insurance company’s maximum settlement value for this claim. (A) 8.89 (B) 9.93 © 12.45 (D) 12.89 (E) 13.53 Lok Xboe|Wro amount OF paagmot doe ~ loynsmral(acs jo =28) HAT 2 einticabor Vasaple oPLA\aher re pasgreuntis reauived RTE vd GO aN oe CETX (DE EWM CX (ZI) —apedtl MoPdam = 0.5 Caeae) Prva J fees = Gay? Tifad P= Gage + Tea? = Tous e.89 Te Expected) Val of 4 chat comonly have tos 2 yobs 1. Th Romo tvs i) 0 us/ Preb 50% it) aay / Probe a7, Veuw (XD) is Bermonlliry with tof 2 vals +i) 0 to poh (2s? b so ‘LE (Ver O (2)Jao.s Corl Cay) )e1se067] Trae \Ja OC = 1506294 19-9579 wn S20¥ Hence = 6.81 to0,02 IHF, F20¥)s 12. YUE. ~~ 4,19, (151-83-94:6] (2 points For number of claims N and aggregate claims S, you are given: Gi) E[S|N= — (i) E[s|N= (iv) Var(S| N= and (vy) Var(S| N=2)=18. Determine Var(S). __{A) 19 (B) 21 (C) 23 (D) 25 (BE) 27__ Vow CS) = Vor CEC SW) + ECVor CSW) AECSIN J. SCBCsw]) r4 EC8=2] = EC za)=y ECECSNT@] = 3 Els JwaJ*+-s ELSIN =2]7= 5 G2 @)=30 ECV ob (sim =4 VorCsin=") Hh Ver CSt/=2)= $a tye) =9 Vax CAG IN) )= ED CSINI*E (ECECS in) 30-44 1y Herc ay CS) = Fly =23 © 4.20. Loss sizes follow a spliced distribution. Loss sizes below 500 follow an exponential distribution with parameter § = 250. Loss sizes above 500 have a density which is a multiple, a, of the density of a Weibull distribution with parameters t = 2, 8 = 400. Determine a. hp yf be the lets si boll oo + Exp (= 250) AX be the des se ohove seo Xa oc POJe Lovbull Ga2 ,o=uce) (X, > 500) sax cdo xihull PCX 9) fi rg Ci-f- oS] wa Fg 7] Lo Fie =q2e = 2.64 5644 ~[S=74a0) = ~~ 4.21. Loss sizes follow a spliced distribution. Below 200 they are uniformly distributed. Loss sizes above 200 follow a multiple of an exponential distribution with parameter 0 = 400. The probability density function is ~~ continuous at 200. Calculate the probability that a loss will be below 200. wane, op 6 4 ~% Unidomn ond. X,~Exp(o= wo) Probab iby Since i € Goo) = POX > 200)= Since pCis cS af acu from PCX{C200) — _tePOXp ren) , ee + LOm)e "a ~ 6 — eo eo _disbubullie. = 250 405 Bo CL rll -a>12pe* see ; 2, =¥, Now Sinfo Pt ae A Ho aus . prlapeDe i erp => p= = 4.22, Loss sizes follow a spliced distribution. Losses below 200 are a multiple a of an exponential distribution ~~ with @ = 300. Losses above 200 follow an exponential distribution with 8 = 400. — Let X be loss size. — Calculate Pr(X < 100). = i 2 Xa f= es belews 2e 5 Xp Fay ») LX, |e lon oo KR, TE = RIELIP CX He ) , aa POX Phow )=SCan) =e ad PCX, 4200) =(\- cents LS scx) NA Waste Dae = © ¥O% 63F U3 t-e to) = ezorese Cine") = o.zoa 224 "4.23, Loss sizes follow a spliced distribution. Losses below 100 follow a lognormal distribution with parameters — 4=3,0= 2, Losses above 100 havea density function which is a times the density of a Pareto distribution with parameters a=2, 6 = 300. Calculate the probability that a los will be greater than 200, _ A~ { jor toys Sizes ae loses how 165; 2 Van ane ob banowel(uas paz) X =a pal of Ruclol La y= So) ATE: Cio ) P CX> 200) Le(X,2 o)-m Ce*) a2 )wlo.2 0, FEE | PCX> wv) = SCico) = saa - @)* 0.5625 PSE, ©,5625 O= 1 > Pleo )] 92. 5° s625a= V— oF 85) S he) fax. Zeoottex, > a=/\|-0, 788! = 6 3767 2.5625 2ePOor»m eal @ |“ 6.574 a0 [horace 2b J Lb aeo4 ~~ 4.24. Loss sizes follow a spliced distribution. Losses below 500 have a density function which is a multiple of an exponential with @ = 250. Losses above 500 have a density function which is a multiple of a single-parameter__ Pareto distribution with a= 3, 0 =500. Half of the losses are below 500. Calculate the expected value of a loss. L on 00 + Eas (o=250 ~b R Co ~3, © =s00 Sop)= je 788 oe Give hak Fls0d) =0,5 Hon V-e azo oO A= —— le over © 30500) <1, Constenh 5 =0,5 is b=0.5 Mean Single aucko = ao > a aa Moar df} Fxpene ok = ‘J re aoe % “I [+ fe chy = SO o> *P250(,-e7%) = 250 -250077 Zia le“? =H HY 12 LE F(x) = 4 160 —0,375e~(—100)/200 3 5 Calculate Var(X). oa Loled Yoluogh a [ns #025 (950+ 1 ZI. 2UI2) = U6o £406 4.25. The random variable X has the following spliced distribution: {3 weigue x f\- ere O

You might also like