0% found this document useful (0 votes)
30 viewsManual Bowers Completo
Uploaded by
Pablo Felipe Herrera JacomeCopyright
© © 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
0% found this document useful (0 votes)
30 viewsManual Bowers Completo
Uploaded by
Pablo Felipe Herrera JacomeCopyright
© © 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
/ 94
yo
Solutions Manual
for
Bowers’ et al.
ACTUARIAL MATHEMATICS:
LIFE CONTINGENCIES AND RUIN THEORY
FOR ‘THE ACTUARIAL STUDENT
Michael A. Gauger, Ph.D.
ACTEX Publications
Winsted, Connecticut
i) Chapter 3
Chapter 4
Chapter 5
Chapter 6
Chapter 7
Chapter 8
Chapter 9
Chapter 10
Chapter 13,
Chapter 14
CONTENTS
15
28
37
41
47
60.
n
83
91
0309595 @ sa = oo|- [non I oo|- [mst]
08 x
Fn) 1—cosx
fa) sinx
© Fe Ine ,
fa) = —
(Note: Age-at-death has an exponential distribution; in such cases, the force of mortality
equals the constant in the distribution, here \ = 1.)
©
s@)= 1 Fa) = (1+ayt
f) = F@) = (+a?
HQ) = 5 = (+a? + tx! = (4x7
Ineach case, p(x) > 0, and [ w00 é& = &
) so -e9| - [rs] ~e*|-
are
i |-e ae |
i
© s@) = oo - [oovar a] = oo| oma]
lo
mee aye
[lL]
s(x) can be a survival function, since s(0) = Landlim s(x) = 0.
k
© sa = eo] [ne a] oo|-she" ep|-2
wr]
= expla inb-alnb+a)] =
FQ) = 1-s@) = al f@) = -s@) = War en Chapter 3
= 121 +x)} 12 < 00
lo
@ [m0 & = “aay ae
hb
(b) Although s(0) = 1 and s(3) = 0, s(x) is not monotonic over [0,3]. That is, s(x) > 0 for
some x.
© [ma [ei de #1 a regia orf). Co prove ti, ety
lo lo
so the integral becomes 2* f yt! e dy, which is 2" (m) > 2,sincen > 1)
s@ _ 1. 100-x _
@ w= SQ” = oO * 10 =
) FO) = 1-3 = 7H
©) = - 40 = Ty
@ Prdo alg = Yo-Ash = -)| = s(0)—s(o0) = 1
cad lo
: 5
9. algo = Pande = ep [-£ wa] {1-00 [fc veal}
= eM — eM) = 00199401 = 002
10. £(65) is a binomial random variable, withn = 10 and p = 5s (65) = .77107. Then the p.f.
ing) = (10) crore 22893), 2 = 0,110
E(£(65)] = np = 7.7107
V[£(65)] = mp(l—p) = (7.7107).22893)
1.7652
11, Letp,i = 1, 2, 3, 4, be the multinomial parameters of the four random variables.
RV. Pi
3Do FR) = 4
sDs FR) - FG) = }
1s FQ) — FO) = §
3Dy F(2)- FQ) = }
(a) Each RV has E|D| = n-p; = :
(b) Each RV has V[D] = n-p(1—pi)
ComD),Dj) =n-Pi: BD 1
Each ri, = a.
© Each ny MD) WD)" PA~Pd 3
7
12, (@) sqo = 1 — 98495 = 01505 sqs = 1 - age ae
fo — fos _ 43180 — 27960 _ 15673
©) ssisa28 — STi1o = 13.
4.
15.
16.
Chapter 3
4
= ak
o | a= & steras ey = PE
af Clearly, if é.¢, is concave down, the drop in the
curve (d,) is greater than the numerical slope, so
best % > wer).
Conversely, if f4, is concave up, then the
mumerical slope exceeds the drop, so. p(X) > qe.
(Note: If 4,4, were linear, then, clearly, w(x) = qr.)
Seen) = be Enea) + way Eby = be we) ~ b wo?
Then clearly Z(G, (2) $ 0 when pay = pay?
§ we ~ nee?)
Both ¥; and ¥, are binomially distributed, with E[¥;] = (1600)
2
Var (Yi) = (1600)(38) (38) = 364, £[¥3] = 40(3%
Var (¥%) = 260(53 = 120. Since ¥; and Ys are independent, then E[¥; +¥:] = 1400 and
Var (¥, + ¥:) = 484, whence SD(Y; + ¥:) = (484)!? = 22. Using the normal approximation,
the value of c, which has a probability of being exceeded of .05, is » + 1.6450. Thus
¢ = 1400 + (1.645)(22) = 1436.19.
‘The future lifetime is t for 0 < T 1.
‘Thus the expectation is 2,57 = Jot sPxpta(Odt +m J-° wPstta(@)dt = fort sPxbis(0) dt +1 mp.
Integrating by parts, 2.55 = —n- prt fo wedt +n «we = fo Prat Chapter 3 5
17, (@) _K*isa function of K so its expected value is given by
mt ent
AK] = SOK WPrK=8 = Sk elart DO mela
7 caf tea
11d
nt ot
= Dk alae te (since srs = > nlar)
ror ea
mt
= Ok Gee eed twee
c=
= (pm aps) + apps) +--+ = aPeaPs) + De
= Pet 22-1) + spB-2) +--+ xpyln—(2—1))
= Petpet tas
(©) Similarly (X*Y? is a function of K and ite expected value is computed as
ot wont
BURY) = SR) PAK = ® = OR alge + OP alas
7 & =
ms
= OP last wee
ca
= Ppa 2p9) + POps-sP0) + PGPe—aPd
te RIP Peat? Ds
= ret ral1) + apg(9—4) 400+ + Pg ——-1P)
= Yek- es
ft
Using the above with Var(K) = E[(K")"] — (ETK")? = E(K"?] — @,77 completes the
proof.
18. (@) Bi7}= & = [est0u wef etd at
©)
E(r} = [sou =e [Pea =e
A f
Then Var(T) = E17} — CET? = 3; -
(©) IfMedian (1) =m, then Pr (T'< m) = 3, 50
Then eo 19,
20.
21
Chapter 3
@ prep [- [vc =ep [- [4]
Petal) =e?
aefe _[Ppectta ai [eett
Oy af t pau dt =i) Pe a=5 [ee at
This is 5 /2r- E{T?], where 7 is unit normal.
Then E{r?] - (E{n)) = aD =1, but £i7] =0, v0 E[T"] = 1
galym = [22
Thus & = 5 2m = 4
ape
If FO 1000-2)", thenf) = FQ = (00-37
) & = AN eae Ao dt [ s000 yt
® %& = EN = : :
‘ lo lo
Loo - 21 al oo
100 a fl af a ,
(o) E{T?] -f[ P00 —x)yt dt = 3(100~xy a = 3(100-2¥
then WO) = ETP (EIN) = 73(100~27
0
© 3= t fo dt =" mip (100—37 ; thus mi) = 3100-9
8 8 ber bl = besabts() = bots = G40) )
Dg Dir er el ey}
C BP OE Ge
Geils) — Ecsale
= fe) = feet puts) - mld)
alu) pat) dt =n) & = 1
©)
(©) Ae = eri — er Nowes = Pr + Pet Pr tos 90
ec = Pall + Prot + Pes +)
Pall + €c41)
Thus Ae, = ean — Pr(l-ters) = ceil =P) — Pe = delat ~ Ps Chapter 3 7
2
2.
(@) _a(a) is the average number of years lived by those who die in [x, x+1]
‘Then a(x)-d, is the total years lived by such deaths. é,41 is the total years lived by
survivors. Thus a(x) -d, + fy; is total years lived in [x, x++1), which is Z,.
() Ly = HG +4) = }(198,740) = 99,370 # Ly in Table 3.2
Ly & Hej +6) = F197,388) = 98,694 = Ly in Table 3,2
(©) This is clear from equations (3.5.12) and (3.5.16).
If s) = 1-3, 0
