ASM Study Manual for SOA Exam MLC Life Contingencies Ninth Edition By I Abraham Weishaus, Ph.D.,F.S.A.,CFA,M.A.A.ALesson 19 Annuities: Probabilities, Percentiles, Recursive Calculations Reading: Actuarial Mathematics ,1-5.3 or Models for Quantifying Risk (3rd edition) 6.1-63, 65-6.6 19.1 Probabilities for Continuous Annuities ‘The distribution function fora continuous annuity present-vakerandom variable Y, in other word the function Fy) =PrL¥ $y), will depend on mortality and interest. Typical patterns are show in Figures 19.1 and 19.2 Lett cscuss these patterns so thatyoul understand them. Fora continuous wholelife ansulty the present vahe ofthe annuity keeps increasingas the annuiten ives longer. However, the present value of later payments is less than the present value of earlier payments. For example, it may take six yeas to accumulate a present value of, bu fourteen years to accumulate a present value of 10. 1°7(x} were uniformly distributed, then in this case F(10)=(14/5)F¥(), so the distribution function ‘rows more rapidly fos higher values. However, 7(2) snot uniform; look backat Figure 4. for atypical pattern ofits density. Asa result, Fly} deceleratesforhigh y. Figure 19.lahas atypical patter. [Nov let's discuss the distribution function of the present-value random variable Yen ofa continuous n-year temporary ife annuity. The probably of Fm < isthe same as the cottesponding probability of whole-fe annuity random variable ify < dp. However, fon can never be worth more than dy no matter how long (2) lives. So the distribution function rises to ngs at y= dm and then jumps tJ. Figure 8.1b has typical pattern fora 5-year temporary life annuity. Bo) BO) 1———. 1——— a8] 08 06) 06 oa a4 2] oa] a (a Wletecen FBS TaL/6=2h Aur tepray We an. Veep ph 0 fers aby = SSE A ta pi FIORE Jo = ote Figure 19.1: Life annvity distribution functions—Part |. Assumes 5 = 0.05, seas iota tion a Copy rASa402, 19, ANNUTTIES: PROBABILITIES, PERCENTILES, RECURSIVE CALCULATIONS BO) BY) 1 1 os 08 05 06 0 04 02 02. y y 246 SDM w mH O27 65 WR 16 1m (3) 5 eer fe cna yO) pq, 01175, and the) 1S yar eran B if any. The vet pat fhe gaph aphiitel at} —dyy= 948738 Ioaiy= gq = 1055967 and (105267) d= 1720. Figure 19.2: Life annuity distribution functions—Part Il. Assumes 5=0.05 Now le discuss the distribution fmeton of the present-vaue random valable »f of continuous n-year defered life annuity. This annuity pays nothing when T(z) $n, and its present values the present value of a \whole-f annuity minus thevalue of an n-year temporary annuity if (x) >. Accordingly, the probability of 0 1s age and F(O)= nde. The graph jumps to nq, st O and aftervardsis parallel tothe graph ofa whole life annuity shifted left by dn. See Figure 18.2 fora typical pattern fora 15-year deferred life annuity. Finally lt’ aiscus the distibution function ofthe present value random variable Ey ofa continuous n- _year certain and life enmity. Tis random variable is never less than dn, because an n-year annuity-certaa Is guaranteed. If7(2)> n, then Veg isthe same as Y, therandom variable fora whole-Ife annuity. So the graph of the dstbution functions up tod, the jumps to age, ten she same asthe graph of to theright ofthat point. Figure 19.2b has atypical pttern of 15 year certain and life annuity Unlike insurance, annuity random variables always increase a a function of survival time. The longer one lives, the more annuity payments one receives. The probability thatthe present valve ofan annuity is less than 2 value corresponds to the probability that Tx) s ess than some value, An algebraic method for relating the istibution of ¥ to the distribution of Tis shown in Table 1.1. Butts ease to season out the relationship for ‘each problem. To reason out the relationship, itmay help to graph ¥ on T to get a better feel for what i going ‘on. The graphs of Figure 193 graph ¥ on T for the four standard sanuties. ‘Most exam questirs willbe on wholelife annuities or defered life annuities. xampis 194 Fora continuous whole ife annuity on (40 (i) ¥is the present valuerandom variaéle ofthe annuity Gi) The probably tour far 4 e0-1)°* inex (85! o Gi) 6=0.05 Caleulate PY $10), Avswer: Based on Figure 193a, ¥ < 10means T'< in tha figure, which means we want gg forthe t satisfying SOME Sud Manes eon Coppice aS191, PROBABILITIES FOR CONTINUOUS ANNUTTIES 403 y y 18 1s] 7 (da) 7 5 5 5 10 15 wt % 5 10 15 aT (Wet sonity (10 yertperesity y y a3} 15 0 o 5| 5 °F 5 10 15 aot a) 15 wT (© Wye dol snr (0 io eeepeay iy Graphs of ¥ on 7, assuming 5=0.05 4n)=10.Solve fort: The desired probability is o ‘As with probabiies and percentiles for insurances, fmotalityis exponential you can skip logging, because youll end up exponentiaing back. IFPr(T'> x)=e°¥, then P(ut < v®)=e"H, and v* =e“, so Pv? < uw, SOA EC Manatees pigeons408 18, ANNUITIES: PROBABILITIES, PERCENTILES, RECURSIVE CALCULATIONS Table 19.1: Expressing Fy) in terms of Fe) ‘Exawtets 19B Fora continuous while life annuity on (40) (i) Ys the present vue random variable ofthe annuity. Calculate Pr(Y > 16). ‘Answen: We need ;pyo where t satisfies én = 20, Then A> 1 te squaeroatt eA, a Exawets 190 Foracontinuous whole life annuity on (x) js =H is constant. Derivea formula for the probability thatthe actcarial present value of the payments willbe greater than the ‘single benefit premium, ‘Answer: We want the probability ofthe event ajpq)> 4, = 1/(u+-6), equivalent to: ‘The probability of this events e-#, oF SOADEC Sy Manal_Be son emp orm AH‘19.2 PROBABILITIES FOR DISCRETE ANNUITIES 403 Quiz 19-1 Fora continuous 10-year defered life annuity on (0; (OY isthe present value random variable of the annuity. G0 pole) =002forall>0. (i) 5=003, Calculate PHY > 5) 19.2 Probabilities for Discrete Annuities For discrete annuities, the solution may begin with the methods discussed for continuous annuities, but then. youmust round the answer toan integer in a way consistent with what the question is asking for ‘Bkawrce 19D Fora whole ie annuity-due on (50): @ Yis the presentvalue random variable @ Mortality follows the tustrative Life Table wm 4 =004 Calculate Pr(¥> 10). “ANSwen"¥ is dggp, and we must compute the probability that this annuity has present value greater than 10 it 004 v1 < 06 nos cue and 6=In(2 +1)=In(14+4/(1~d}) =In(1+1/28)= 0.040822. $0 K +1 >~(in0.6)/0.040822 K 212, the annuity’ present values greater than 10. Prom te lustatve Life Table, fea _ 7.954179 iba 95090 Frequent, exam questions ask forthe probability that the sum ofthe annuity payments not discounted at Ancarest ae greater than the single benef premium. Eustis 196 Fora whole if annuity immediate on ($5): © poxlt)= 1160-1) for-10 2.51. So when po A tev 1g? * spo "3 10315237 a= Tene 7205 455 = 12.0875 —1= 1.0875 sonnet Manson ep ets AS408 19, ANNUITIES: PROBABILITIES, PERCENTILES, RECURSIVE CALCULATIONS are greater if there are atleast 12 payments, or if ($5) survives 12 years. But pes = (50 ~ a 9 (Quiz 19-2 Fora 10-year certain anlife annuity due on 60 @ palt)=1140~1)for0- 15.1969 1/0 Toe? 1ssea9 1-(1)' sosores 104, ‘Loe > 2.58011 ¢In.04> 255011 12> 73.8684 ‘Therefore, (5) mst eat ea ero he th ihn. a SOAMC Su Maal eon onyx A408 19, ANNUTTIES: PROBABILITIES, PERCENTILBS, RECURSIVE CALCULATIONS (Quiz19-8 For a 10-year temporary life annuity-due on (40), ( Yisthe present value random variable, i) Mortality follows deMotvre’ law with « = W 1=0.06 Calculate the 25th percentile ofthis annulty. 19.4 Recursive calculations ‘We can relate annuities atage.x to those at age x++1. The following states the rules for the four standard annuities with thee versions aplece: anmuities-due, annultes-Immediate, and continuous annuities. However, itis rare ‘that we use recursion on continuous annuities. Wholelife annuities For an annuty-due d, is dzy1 discounted one year, with an Immediate payment of 1 added, 30 A= 0Pedg th on Foran annuity-immediate a, i a; discounted one year wth payment of atthe end ofthe year added: Oe= WPedssi tb Fora continuous annuity, ds is dvs discounted one year with dy added: y= vpsdeart den ‘Temporary life annuities For an annuity-due, dion is é:4.7-7 discounted one year with an immediate pay- ment ofl added: Gem = UPd sent] For an immediate annuity azn Is 2,7 discounted one year with a payment of at the end ofthe year added: em UP Oya sent UP continuous annuity dein is d;,xa-n discounted one year plus a temporary 1-year continuous life For Gem =UPedynentaen, Deferred life annuities When bringing these back through the deferral period, the next year’s amount is dis- counted, but nothingis added, mide = UP ga yeat (Oe = UP gash nde= Ps gy deer nyear certain and life annuities During the certain period, there is an immediate payment of 1. If death oc- ‘curs, theres still an annulty-certain benefit. The formula for an annuity-dueis aya tug dent UPd ‘SMC Sn Naas B eon cpp 208 ASEXERCISES FOR LESSON 19 409 Foran annuity-immetiate the formula is 0g. ant OP Fora continuous annuity, the formulas int vedanta Recursion formulas are useful ifa change ismade to an assumption. ‘Baars 191 For 10-year temporary ie annuity on 3), youre given @ ge=0.01 i #=007 @) agm=5 Calculate the change in the actuarial present value of this annuity due ifg, is increastd by 03, ANSWOR: dgsq= 14+ UPedge1, 50 If we change qu. the revised present value is 096 14958432329) =4.07079 + pplaaese) ‘The change is therefore 4.87879 ‘A whole life annuity may be spit into an n-yeat temporary annuity plus an n-year pure endowment factor times an n-year deferred anmuity: and we used this formula in previous lessons, ‘There are recursive formulas for increasing and decreasing annuities as well. These are similar tothe ones ‘we developed for increasing and decreasing insurances in Section 14.2. For example, (la\,= a+ op Aan ‘You should be able to develop such formulas as needed. it 5B gle Exercises Probabilities 19, (250-898:15| For a5-year certain and life annuity of | on (x), payable continuously, you are given: (Vis the present-valuerandom variable. i) Fristhe distribution function for ¥. Gi) pAt)=001, 120 u) 6=008 Calculate Fy(30) BS). won (@) 088 ost ©1085 © 100 SOA aS ton ‘erie conti onthe nese Cony aa ASa0 19. ANNUITIBS: PROBABILITIES, PERCENTILES, RECURSIVE CALCULATIONS Table 19.2: Summary of Formulas and Concepts for this Lesson Probabilities and Percentiles + To caloulate a probabiliy for an annuity, calculate the for which d; has the desired property. Then calculate the probability ¢is in that range. + Tocaloulate a percentile of an annuity calculate the percentile of T, then caleulate dp. + Some adjustments may be needed for discrete annuities or non-whole-life annuities, as discussed inthe lesson, + fforces of mortality and interest are constant, then the probability thatthe present value of payments ona continuous whole life annuity will be greater than its actuarial present value is Plea +1 Patass ts Peden +40 Paden gat) Gem VP Bsuzy+YPs eg = UMC OUP OH 4en=Pebonsent det aaqn ag tigen Upc dsraey 19.2, J250-$88:12] A10-year deferred life annuity due of 1 per annum is issued to (50). You re given: (Mortality follows de Moivre law with = 100. w i=0 Calculate the probability thatthe sum ofthe payments made under the annuity will exceed the actuarial present value, at issue, ofthe annuity a) 04s 048 © 080 (D) 052 @ 054 19.3, {150-891:12] Fora 10-year deferred life annuity ge). ost @ 082 oss 1D) 0st ® 085 419,10, Ys thepresent value random variable for a continuous 10-year temporary life annuity. You are given: 2 for all x. Calculate the probability that otal payments on this annuity wil exceed E(Y}. 19411. Ys the present value ofa whole Ife annuity-due on (20). You ae given: (© Mortality follows de Moivre law with «= 100 Gi) F=006. Calculate the probability thatthe sum ofthe payments will exceed EIV]. 04a sety mane —ooeon arcsec on he est pe Ceoniicoens astEXERCISES FOR LESSON 19 413 1912, Kira purchases a3-year deferred wholelie annuity with continuous payments, You are given: Calculate the probability thatthe present value of the payments on this anzuity wil exceed the actuarial present value of the annul 1918-16. Use the following information for questions 19.13 through 19.16: (i) Mortality follows De Moivre’ law with a= 3918. [150-898:31] Calculate “As. (A) 0.1536 @ 01897 (© 0.1648 01701 @ 01785 19.14, [150-98:22) Calculate ds (a) 14.08 @) 457 (C1496 ) 1535 @is74 19.15, [150-898:33] Calculate PX Ary > ds) w)0se @ 059 088 ©) 048 069 19.16, (150-898:34] Calculate Var) wa @) 26 ow ©) 30 ox 19,17, [280038] For continuous whole life annuity of Lon (x) (Tx), the futur lifetime off) follows a constant force of mortality 0.06, (i) The force ofinterestis 0.08, Calealate P(dgy > a). ww 040 04 (© 045 0) 048 © 080 ¥8.18._[CAS4A-F98:8] points) You are given that(x) is subject oa constant force of mortality, z= 0.06, and thatthe force of interests also a constant, = 0.4, isthe time until death random variable. Calculate the probability that dy > 1.504. (Less than 0.20 @) Atleast 0.20, butles than 0.40 (©) Atleast 40, butles than 0.60 (©) Atleast0.6, butless than 0.80 Atleasto.s0 ‘04n0C uty Manuals eco rrcescominueon he eet page. eyecare nS44 19, ANNUTTIES: PROBABILITIES, PERCENTILES, RECURSIVE CALCULATIONS 19.19. [150-F97:26] You are given: o (ii) Ts the futwe lifetime of 2). Calculate the probsbilt that dp al exceed dam, (a) 070 won (0a most 6) 098 Percentiles 19.20, Yisthe present valve random variable for a continuous whole life ammuty on (60). You ate given: (Mortality follows de Moivre law with o = 110 @ 5=006. Calculate the 90th percentile of Y. 19.21. Youare given: Ys thepresent value random varlablefor a whole life annuity-due, Gi) ge=0.1 forall. (iy 1=005 CCaleulate the 60th percentile of Y. 19.22, [CAS4A-S96:12] 2 points) Alife, age x, wants to fund alife annuity witha deposit of amount P, This it {is subjecttoa constant force of mortality, = 0.06. 5 =0.08. Determine P such thatthe probability is 85% that P willbe sufficient to fund a life annuity of $100 per year payable continuously (Less than $1,000 (6) Atleast $1,000, butless than $1,100 (©) Atleast$1,t00,butlessthan $1,200 (D) Atleast $1,200, butess than $1,300 Atleast $1,300 19.23. Youare given: @ e001 forallx. @ 6=0085. Let ¥ be the present value random variable for a 5-year deferred continuous life annuity. Calculate the median off. SAN ond Mana -60 oon veces continuo the net page ony ox[EXERCISES FOR LESSON 19 415 19.24, 20-year deferred whole life annuity on (x) pays I per year continuously. f(x) dies within 20 yeas, the single benefit premiuin for the annuity with interes is refunded. You ae given: @ w(t) =0.02Forall > 0. ( 6=006 (li) Ys the present value random variable forthe annuity Determine the median of Y. Recursions 1926. [3-S00:29] For a whole life annuity-due of 1 on (x), payable annually g=001 gen =0.05 Calelate the change nthe actuarial presen ue ofthis annuity. dei pis Increased by 0.03, 8) 036 @) 017 ora 029 020 19.28, [160.5885] You are given @ d=6forallintegralx. (i) f= 008 Calculate ws. (ay 0268 @) 0364 (0587 ©) 0636 @ 0737 1927, [¢AS4-$86.19] (1 poin) You are given @ 4 agus (i) Som =6 Caleulate Ey. (A) uessthan 0.75, (B) Atleast 0.75, bat ess than 0.80 (©) Atleast 0.80, Butless han.085 (D) Atleast 0.85, butless than 0.90 ©) Atleastogo SOAMELC ut onal eon urs contin tena poe. Conyac oasoasteANNUITIES: PROBABILITIES, PERCENTILES, RECURSIVE CALCULATIONS 416 19.28. [CASAA-$99:17) (1 poiny You are given 0 15.85 cy 3.45 iy i= 0.08 Calculate the actuarial present value of a 5-year deferred whole ife annuity-due issued to (30). (A) ess than 100 (B) Atleast 10. butless than 10.5 (Q Atleast 105, butless than 11.0 (©) Atleast 11.0, butless than 11.5 Atleast 115 19.28, [CAS4-S84:21] An annuity is ssud to Sally, who i age x. a, isthe actuarial present value of an annuity assuming standard morality, but Sells mortality inthe n-+18t year fq, 1 €QUa tO dren +, Where doom 18 standard mortality. Which ofthe following expressions mustbe added to ato obtain the actuarial present valu of Sal’ an- nuityt (A) C0" aPe-dsonet Co" nPedasne ( ~cv" aperaenn, ) ~C0" np tase (B Thecorrect answer isnot given by (NB, (), or (D). 19.90, [CAS4A-F52:21] 2 points) You are given the folowing fe table data: x bk ge de 50 508 51 9.00800 52 91,865, ‘Yu are given that f= 0,08 and asy =11.888. Determine ase. (A) Less than 12.083 (B) Atleast 12.083, butless than 12.89 (©) Atleast 12.008, butless than 12.095 (D) Atleast 12.095, but ess than 12.101 © Atleast 12.101 Sans Mana8redton Bae continueon he et page. ‘epi 2c ASK[EXERCISES FOR LESSON 19 a7 1931. [CAS#-S85:18] 2 points) Given: dy=25.08 Gi) ay =24.85 a) i008 Calculate pis. (A) Less than 0.9900, (B) Atleast 0.9900, butless than 0.9825 (©) Atleast 0.9925, but les than 0.9950 (D) Atleast 0.9950, butless than 0.9975 ® Atleast 0.9975 For an annuity-immediate on (x), youare given: 0.012. Themortalityrate assumption di then changed to 0.00. Caleulatethe revised value of ax. 19.33, The actuarial present value of § year deferred continuous whele ie annuity evaluated at 5 = 0.06 is 10, The interes assumption forthe fist 5 years ony i then changed to 6. Determine the value of for which te actuarial present value ofthis annuity is 102. 19.94, For(s), you are given ‘The mortality assumption for age only then changed tog, = 0.0. Calculate the revised value of 5, 19.38, (150-F97:14] Youare given: © usp. + opeueetNx=0,1,2,.. ®_u(6s) Which ofthe followingexpressions is equa! to u(40)? (8) geoam ©) ars © dnsn ©) doz ©) The correct answer is not given by (A), (8), (©), or O). soAnacuty wants ein ures continue on he nex pe. Grorebeene eoa8 19. ANNUITIES: PROBABILITIES, PERCENTILES, RECURSIVE CALCULATIONS 19.96, {SOA3-Fo4:29] At interest rate f @ a,=56 (i) Theactuarie! present value of 2-year certain and life annulty-due of 1 on (x)is dy = 5.6459, Gi) eg =883 (i) ex =8.29 Caleulate ¢ (0077 @ oor (@ 0081 ©) 0.088, 000 ‘Adiiional released exam questions: SOA M-S0526,CAS3-S0621, SOA M-P06:33 Solutions 19.1, Wemust calculate the probability thatthe present value ofthe annuity is between 5 and 30, or PAS< ¥ $ 30) =Pi(¥ < 30)—PHY <5). Note that a perpetuity would have present value } = 25, so the present val alwaysless then 30, and PY <30)=1, Hence wenesd A-PH(.S5)=PHY>5) ‘The annuity would have present valueless than 5 if death occurs during the certain period, so the probat of the present value being greater than § is less than 1. We need the time ¢ such that the present vake of annuity is exactly 5, orf such that n=5, and then the answer willbe ;pe =e? lt oe 16% =56=02 oom 08 =o 19.2, With an interest rate of ze, the annuity-dueS expected present value is the expected sum of ts pay- ‘ments, If 50) lives less than 10 years, that is 0, and if (0) lives more than 10 years, the average number of payments is 205, since it ranges uniformly from | to 40, so the average is the midpoint or 20. Therefore, the ‘expected sum of the payments is Pe 1040} wpse(205) = 0.8205) To exceed this, 17 payments are neseied, To get 17 payments the annuitant must live beyond age 76. The probability ofthat, ap», is [GAB (8) 19.3. Vitually the sameas the previous question With an interest rate of zero, the annuity due’ expected present values the expected sum ofits payments 1 (60) ives less than 10 years, that is 0, and if (0) ves more than 10 yeas, the average number of payments i 155, since it ranges uniformly from 1 £080, so the average isthe midpoint or 15.5. Therefor, the expected sum ofthe payments is sofea(O) +9 peel 18.5)= 0.75115 5)= 11.605, ‘To exceed tis, 12 payments are needed. To get 12 payments, the annultant must live beyond age 8. The probability of that, 2 peo, is 19/40 = [04751 (A) SOMES Marken ‘ep 6208EXERCIS3 SOLUTIONS FOR LESSON 19 419 194, There sno need to solve this precisely since the correct answer can be arrived atby elimination, The probability Y < 10,000é7q = 63212 is, which eliminates (§, (Q), and (2). The probably Y = 1000049 is the probability oflivngless than 10 years, which sn 0, and this eliminates (B) which doesnt have a jump at $$63212. A perpetuity would be worth $100,000, s0 F{100000) must equa! (D) Ifyou are mathematically inclined and want to understand how to derive the right curve, the solution is as follows. Consider the variable ¥"= ¥/10,000. (We use Yt simplify the formulas) If etme to death the valu of Yis dy =63212:E Tis less than 0, otherwiseitis dm =(1—e°**7)/0,1. The higher T's, the higher ¥is. So FQ)=PAY’sy) =n(io(1-e") 1107, Rather than solving a quadratic, it's easer to do this by estimating K2 > 2(1107) 0 > 47, Trying k = 47 gets (47\(48)/2 = 1128, o 47 payments exceeds 1107 and 45 payments doesnt. So the desired probability is wpea =(80~ 45)/80=[is425 Sonnac Sn Manas een epic e295#20 19, ANNUITIBS: PROBABILITIES, PERCENTILES, RECURSIVE CALCULATIONS 198, The actuarial present value ofthe deferred annuity is by the current payment formula with = 1, ‘The probability of more than 82 payments isthe probability of surviving 102 years, o @"Pt 198. By formula 07.7), reget Either way, gs =20.621 -4.542= 16279, To get 17 annuity payments, the annuitant must survive21 years. The probability of thats ap, —e-*! = [ED () 19.10. EW ‘The probability of payments over a period of atleast 7.519806 years ise" eHH7 189 19.11,_ tis asiero calculate the expected value of Az and then to calculate i from it A 1 fa ama rat (seas) (5) 1011782(17.49968) = 0206964 ‘Ay _1~0.206364 _ tae = Spans = 1208 fo ‘The probability of 15 payments{s the probability of surviving 14 years, or 1 ~ 14/80. 1912, The actuarial present value ofthe annuity is ene a 921/007 = 115798 eae 0.07 = 11.579 We want the for which a deferred continuous annuity equals 11.5798, and then the answer will be¢pe =e"? SORES Manes eon oprit eastEXERCISE SOLUTIONS FOR LESSON 19 a2 19.13, As usual with de Moives iaw, this is an annulty-certain for period w ~x divided by «—x. a 19.14. Easier to calculate Ass and take it from there. = Sess A= 2 a-et) 1-0.295762. a 005 1915. We need t such that én = aq 14.0848 (rom the previous problem), and then the required probability will be pss. tlny=In0.29576=—1.218207 ~12is207 _ 1.218207 inv 43612 Payee =1- 2388, i © 19.16. From the previous problems we already have the first and second moments for a whole life insurance, sowellluse # 01896-0255762 pe i BES] © 10417, Foracontinuous whole annul, we derived in ample 19C thatthe probebillyis wey oo6 yi F (cts) (oes) = [S888] We want the probability ,p,, where ¢ satisfies: 1918. &,=1/tU+6)=' leet 5 leet 004 e204 ‘The probability is 6 SOANE Sty Minat-Preion opments a422 19, ANNUITIES: PROBABILITIES, PERCENTILES, RECURSIVE CALCULATIONS 1919, First we calculated, Weneed Pr(T'> #) = e°29 0 =-9,417662°* oO 19.20. The 90th percentile of 7 for this uniform distribution on [0,50] is 45. The present value of a 45-year continuous annuity is 19.21, The survival probability ,p,=0.9, We want .q,=0., s0set ep equal to 1 ast=o5 nod k=O wa.so6r2 ‘The 6oth percentile occurs when time to death s 8.62672 yeas, in which case there are 9 payments (mes 0 ‘through 8). Then 4g = (1 (0/1088) (0.05/1.05) 19.22, Wewantthe9sth peroenleof T(x). We want Such that z= Pwillbe oo Peon, ==). 008 owever since e-** =0,06, ifolows that e-P@ = (e~18) panzsot-005") =] 0) 19.25, From the fact dat jg = 0.01 forall x, PAT > ¢) = yPy = eH, We want to detenmine 1 such that Px(T> €)=05, or e*!'=05, Then well calculate the median of Vas — an = 55 However, e808 = (2-201)!4 9 2080 0585 = 0.044194, 0 the medians 9250044194 _ 0.798516 -0.044194 0045 0.085 SOAS Std Man _sheion Cony ensEXERCISE SOLUTIONS FOR LESSON 19 423 19.24, Let Tbe survivaltime, The present value ofthis annuity is lowest for T'=20, then grows as T gets larger. However, forthe right 7, ithas the same value as for T <2. For T < 20, the present value ofthe annuity isthe single benefit premium, which wel cll P. Lt calculate P. We equate P tothe probability of dying in the fist 20 years times P, plus the probability of surviving and getting @ deferred annuity. P=ngdsP+¥" apederee etm tose) 1 1 8 Cero e-s610100) = tay — (0 P+ oe Lets calculate the probability of surviving until the time 7 for which the defered annalty s worth P. The de- ferred ennuity is worth Setting this equal to 3.76493, 0.063.76493)= 0.225896, e7!3—0,225896=0.075299 In0.075299 006, 43.1089 ‘The probability of surval to T= 3.1049 is e-tH9 = 0.422275. The probability of survival to T'=20 8 0220 = 0670320. This means that the probability of survival to atime in 204.1049), the times for wich the presen value oithe annuity is minimized is less than 05 (tis 0.670320 0.422275 = 0240045 to be exact), 0 these tines are not the median. On te other hand, the probability of surviving longer than 43.1049 sess ‘than 0.5, so the median snot greater than 43.1049 either. That eaves survival umes less than 20, for which the present aueof the anny isthe sage preniun ETOHES. That the median, 1925, Weuse primes for therevsed values. ‘The original dy is 1 +(0.89/1.05(6.951) = 7.5528. The revised difference is 7.730988 - 7.5538 =[,0.4722 | (C) 1926. Wecan relate to dz. with recursion, then use that theyre both & 14 (099/1.05)7.13999 730988, The a= 14 0pden fae Sane Maal_steten Coppin ons Ae424 19, ANNUITIES: PROBABILITIES, PERCENTILES, RECURSIVE CALCULATIONS p28 case sto=1~0948" > ® 19.27. Wespltthe whol if annuity into temporary and deferred: 4.2 Aart gayle 1am gaye Mal an+18 axm=5 However, asin = 0B Sem 19.28, ‘The interest rate is extraneous 19.28, asm tv ypall dean + dst) reat V gpa aha gXl-F ase) The difference is ~co™!,ps(1+ aryns), whichis 19.30, 00=1P.0+ 9p 1 = 28-9165 lg =9L9165 4508904045 n= ES oss Se + 1.8) i203] © 1931. ay= 14 0pisdn 14 29, 2sot=1+ P2405) (2404)0.03), Pio= agg = SNE o 0A. ud Maas eon enpii 2eASKEXERCISE SOLUTIONS FOR LESSON 19 425 1932. Weuse primes for revised values, 18.4=0.96 + 0.960(0.98841 +442) 0.96y(0.988)(1 + azya)= 1744 17440981) oom 3.96 +1740 0.960(0.991)(1+ az42)= 19.3. 920918003 030019003 7 19.34. We use primes for revised values. We can express the annulty asthe value ofthe curent payament of 1 plus the values ofthe certain annuity if death occurs or the 9-year-certuin and life annuity if survival occurs. i tugs a+ Up teaay 093! 17035 -$(0.95)(0.01)7 395012) +-(095)0.99)4 +5 aT 070259940505 ay =7395012 b= =14021076+ 13.6834: Alternatively, we can express the annuity asa 10-year annuity certain plus a defered life annuity, and use recursion on the deferred annuity: roids =15~ 8025261 =6.974739 wif Pages = (0954038) 44su1 ag7739 (0.95%0.99) .95X0.97 (7 41599) = 6.833835 casa +-0030005- TE 41599 aoe SDAA. Su Maal -96 eo ‘Compson426 19, ANNUTTIES: PROBABILITIES, PERCENTILBS, RECURSIVE CALCULATIONS 19.35, We start with no present vale a 65, Going back one year, we discount the amount and add on a pay- ‘ment at the end ofthe year if the life survives (vps). Hence tis annuity wil pay one a the end ofeach year up ‘to and including the year the life turs 65, since at 6, u(64)=vpe, So the answers (A) 19.36, Fist we note thaty the recursive equation fore, equation (73) on page 106: 8.83=p,(1+828) 803, P= m9 (Comparing aie annuity to 2-year certain and life annuity, we ind hat othe differences v(1~p,). Then we have Quiz Solutions 19-1, The actuarial presen value ofthe payments if death occurs at ¢ yo—ot and we want the making this equal to 5. The answer willbe ,>ro v®— »#=0.03(5)=0.15 vfs ul 0 9.590818 3 9.15=0590818 ‘Then e-*s 0.508188? = [O70Ai6) 19-2, We have dizay = dio mgd. Since mortality is deMoivre, well cleulate th defered annuity using insurances. 1/089) ~-40(0.05) S0AnEC uty Mansion Coprie s it(QUIZ SOLUTIONS FOR LESSON 19 ar vadeg= OSSA7A~O428977 yoy 005/105 ‘ ‘The 10-year certain annuity has actual present value 1~(1/1.05"9) a= Toes O07 ‘The actuarial present value ofthe 10-year certain and life annulty-due Is 4.71452 + 8.10782 = 12.82, Therefore, 13 payments are needed to top this value, which happens when the annuitant survives 12 years. In other words, the answer is gpa =(40—12)/40= fie 19-3, The 25th percentile of Tis 60. The annuity assumes its maximum value at 60, but that doesnt matter; digg Is any percentile fromm 125th to 100th. gg 2 Brtw/.06 m= D066 ‘saAa.csely sana sean ‘en eznsasie328 419, ANNUTTIES: PROBABILITIES, PERCENTILES, RECURSIVE CALCULATIONS Ona. sty Maal -Bh eon op aseLesson 20 Annuities: m-thly Payments Reading: Actuarial Mathematics 54 ot Models for Quantifying Rsk (3rd edition) 64 This topics rare on exams. You wll ecognize questions on this topic and te teated ones for insurance and premiums because they always mention tat deaths are uniformly distributed over each year of age. ‘Amodal annuity ays /m every mth ofa year. Its actuarial present values denoted by 2” fora whole fe annuity due, with the usual modifications for other annuities. The differences between an annuty-ie and an ‘thy annulty-due are: a 1. The mthly annity-due will make its payments att. Te value of delaying payments i aie annuity of the difference between a -year mthlyannulty-certin anda paymentof I, or (d/d(™)~ 1), 2 The mthly anuity-due wilt pay asthe mthly period of death nthe year of death. ‘Atraitional approximation forthe APV ofa modal annuity due in terms ofan annval annulty due is m-1 — eo) ‘which is correct if i = 0 anid deaths are uniformly distributed over each year of age. Similarly for immediate annuities, ae. Actuarial Mathematics derives theve formulas under a somewhat mote general assumption, namely thatthe expression v**,44pz is linear ins forall x and 0.< $< 1. However, these formulas dont work well with high interest rates and low mortality rates. To drive formulas that are correct for uniform distribution of deaths over each year of age, regardless of interest rates, we define ia . oe) = Fagin faim Ble) = Sega Both of these functions depend purely on, noton the year of age or mortality at. The formulafor the APV of ‘2 modal whole life annuity-duein terns ofan annul whole fe annuity duels ay (202) 1m corresponding formulas fr temporarylife and deferred life annuities, f(m)lsmullipied by 1 minus the pure cendoviment o by the pure endowment alon= ofm)em— (mY ~nEe) 203) A= aim) de Alm aE 04) ‘To modilize annuities immediate, um the modal annuity into modalannulty due use the formulas above, then tun the annual annuity due into an annuity immediate. For example, a”=al")~ 1/7. {A{m)is the cancelation coeficient for the payments not made after death. In other words, it represents the ‘expected accumulated value a the end of the year ofone year of payments canceled by death, For example, 2) ‘would be the expected value of sos nals etaen 429 Cepreic eae ast430 20, ANNUITIBS: m -THLY PAYMENTS + 0.544995 when death occurs in the fist ha of he year, since then the 05 payment at months not made. ‘+ Owhen death occurs in the second haf ofthe year, since then both semiannual payments are made. Under uniform distribution of deaths, the probability of death occurring in the first half of the year is 05, s0 Smiley, 6(9)=0/9)(04 8994-20 +78), Tyou didi want to memorize the formula for am), you could use the formula. =) (aa) ~plmAs e(m) willbe 2 tle more than 1 and Am) willbea lite more than (m —1)/2m. Ifyou caleulate these and dont get values in ths range, check your calculation. For whole fe annultes and temporary hfe annultes, rather than using these formulas, you can convert whole if annus into whol lif insurances o temporary lif anmutes into endowment insurances, use the easier insurance conversion factor (i/1(), and then conwvert back to annuitles, Remember that for endowment insurances, the 1/ factor is applied only to the term insurance, not the pure endowment. The conversion formulas for converting the modal insurances back nto annulies are the same asthe formula for annual in- surances and annuities, except that you use the modal interest and discount rates. Fr example Aim=1—amate) ‘The tables you gett the exam have a{m) and A(m) at 6% for m=1, 2, 4,12, and oo So you will usually not hhaveto calculate a{m) and f(r). Exawp.e 20 Assume uniform disuibution of deaths between integral ages. Calculate alt) based on the Iustrative Life Table at 6%, ‘Answer: Welook up the tables and see that a4) = 1.0027 and f(4)= 0.38424. Als, dss =12.2758.S0 af =olA)ts- B14) = 1.00027(12.2788) 0.3082 Using the simple approximation of equation (20.1) would yield 1227581 ~0.375 = 11.9008, which would probably be good enough onan exam. ‘odo this fom the insurance formula, i 006 7 0.05870" 1-a@_ 1031190 ag (0:30514)=031190 er S Exercises 204. [Based on 4-Fo615) You are given that deaths ae wnifomnly distributed over each year of age. ‘The function iq t'hy equals a(12)a4s5im1— BU2)E. Determine k. (A) awEas (B) 2B es (C) t-Eas. (D) 1-20Es (B) 1ozs ~ 20 Fas SoaNaCSny aa eon ri contin onthe net poe ‘ephc obASEXERCISES FOR LESSON 20 tt ‘Table 20.1: Summary of Formulas Relating Modal Annuities to Annual Annuities Under Uniform Distr of Deaths ap") = am) — Bm) (202) aS) = alm) ~ Blo KL~ a8) wa) alm) ys— BeBe 2 ea) where 20.2-3. Use the following information for questions 202 and 20.3: 7p caeulate the values of annuities with m-biy payments, youre to assume that gape olinearin foralls and0S¢¢1, You are given: 0 dgq= 150 a= 145 0 20.2. Caleulate ay, : my 20,3. Calculate a, 204, Youare given (@ Deaths are uniformly disibuted over each year of age. w Gi) w o 2, Calculate a2, 205, Yousre given (0) Deaths are uniformly distributed over each year of age. Gv) 5B, =0.20 ot SoANaC Say Mans eon ‘rer continao the net poe. ‘epi czas432 20, ANNUITIES:m -THLY PAYMENTS 208, Youare given @_ Deaths are uniformly distributed over each year of age. ) AP =o, (i) 42, 0.2125 ) Arg ate a Catealte a, 20.7, [150-St8:24) Youare given: © Aln=o0ui9 GH) Be = 0.54733 aio 05 Gv) a(4)= 1.00019 © AL ‘Assume deat are uniformly distbuted over each year of ag. Calculated, (A) 883 (B) 9.00 (© 9.04 (0) 9.10 (©) 9.38 20.8. {150-81-94:11} Which of the following is tru, regardless of the assumption about deaths within each year of age? MW. a,=4,9+5 Peden Ml. a, =aloo)de~6(0o) ™ Ap=1-6a, (A) None (B) Lonly (©) Honky, (©) Hlonly ©) Wonly 209-1. Usetefellowing ijormationforquesions 209 though 20.11: [Taoeegee @ 05 i) Deaths areunormly dtibuted ove ech yar of ag. | Gi) ay’ = 14.500 | 20.9. [150-82-94:34] Calculate A (a) 0.253, 8) 0.264 (©) 0277 (0) 0.288 © 0299 20410. [150-82-94:35] Calculate a. (a) 14.75 @) 15.00 (© 1525 (D) 1550 © 15.74 20.11, [150-82-94:96) Calculate A. ) 028 @ 029 (030 031 ose saad analstheion Bares cortnur on the net pe. Copriacone astEXERCISES FOR LESSON 20 433 20.2, [Based on 4-FH6:18) You are given 0 a 0.5688 59.9259 5.9348 235516 Ta] tw) Deaths are distbutd uniformly over each yar of ag Cateutte 1a, (A) 27 B) 28 (02 @) 30 20.13. [150-P96:19] You are given: deta a Ar=oa0es (i) Deaths ate unormycoubued over each year fags Calculate a, a) 15.48, @ i551 (@ 1535 ©) 15.82 20.14, [S0A3-P0St0] Foran annuity payable semiannually, you ae given: (Deaths are uniformly distributed over each year of age. . ass a7 wy aii) (Gv) Deaths are uniformly distibuted over each year of age ulate a, Calculate a ea @ 1586 @ a5 20.16, Aspecial lif anpuity-immediate on (50) pays 10 per year on each anniversary starting wth I year after the contracts purchased and 4 per year § months after the anniversary starting with 6 months after the contract is purchased. ‘You are given that i = 0.06 and mortality follows the Illustrative Life Table, with uniform distribution of deaths between integral ges. Calculate the actuarial present value ofthis annuity. SOMES analSh eon ‘Conyac omensaf =alh2)ais- (12) 2)as5— 02) 12)a45— (12) vndayan = 108s ssn es and he last equality holds when is replaced with in al’ postions. ais that the answers). 202. With thelinearhypothess and m=2, wehave(—1)/2m=025 and 0 that Bs =0.8, Then 23. dyed (Jona ara Aen” 4 (g] OH = 1425-025 +025(0.8) =f) 204, With #=0 and uniform dlstetbution of deaths, the formulas using a and fi do not work, However, the simple approximation (forms (20.1) is exact inthis cas, s0 atin Overs you cou waite fom it prinpes Shee =0,thepesentrae oft anntys sum cttapnymnt Winns fot detour Thowe wo den et month get 2! less with the monthly annuity; those who diein the second month get Jess; et. So the difference Tereon inmates eth abi dg tetwem drsons and 15 was OSD” Ooo 1ee wien 0 a8oapshemes a (Fe alt ‘Thus we get the above solution. 205. a(n") (o7)/1.1 (0.0956897{0.0949327) 0.1 -0.0956097 .09568970.0949327) 8, = (00078045. ~0.47991(050 020) = ETAT = 1.000752 0A. dy Mane olon ‘epi ozo