450 likes | 897 Views
Pengantar Matematika Asuransi. Winita Sulandari. What is insurance mathematics?. Insurance mathematics is the area of applied mathematics that studies different risks to individuals, property and businesses, and ways to manage these risks. What is insurance and w hy do people need insurance ?.
E N D
PengantarMatematikaAsuransi WinitaSulandari
What is insurancemathematics? • Insurance mathematics is the area of applied mathematics that studies different risks to individuals, property and businesses, and ways to manage these risks.
Asuransi • Asuransi adalah sebuah sistem untuk merendahkan kehilangan finansial dengan menyalurkan resiko kehilangan dari seseorang atau badan ke lainnya.
Istilahdalamasuransi • Polis • Insurer • Insured • Premi • Klaim • Risk : averse, neutral and loving/seeking Perjanjian antara kedua badan Penanggung: badan yang menerima resiko Tertanggung: badan yang menyalurkan resiko • Biaya yang yang dibayar oleh tertanggung kpd penanggung • utk resiko yang ditanggung tindakanuntukmemperolehhakataskerugian yang diderita olehseseorangsesuaidenganjanji yang terteradi polis asuransi
Jenis asuransi • Asuransi jiwa • Asuransi kerugian • Yang membedakan adalah pada obyek pertanggungannya
Asuransi jiwa • Asuransi jiwa dalah perjanjian tentang pemberian santunan dalam jumlah tertentu yang berhubungan dengan hidup matinya seseorang • Tujuan asuransi jiwa adalah perlindungan pendapatan (sementara) keluarga (ahli waris) nasabah
Asuransi kerugian • Asuransi kerugian adalah asuransi yang memberikan ganti rugi kepada tertanggung yang menderita kerugian barang atau benda miliknya.
Bagaimana dengan asuransi beasiswa? • Apa bedanya dengan asuransi jiwa?
Asuransi beasiswa • Asuransi beasiswa adalah asuransi yang mengcover resiko biaya sekolah anak, apabila terjadi resiko pada si tertanggung.
Effect of insurance • Example:
Effect of insurance • Prob of loss is 0,5 • Expected level of wealth: 90.000 (0,5 x 100.000 + 0,5 x 80.000) • Misal : Asuransi 10.000 denganharga 5.000 premium = insurer’s expected claim cost = 0,5 x 10.000
Effect of insurance • Example: • * with insurance
Effect of insurance • Perhatikan bagan berikut • Asuransi wealth (LOSS occurs) wealth (No LOSS occurs)
Risk Aversion, neutral and loving Example: Expected payoff = $50
Risk Aversion, neutral and loving Risk attitudes: • Risk averse: Accept a payoff of less than $50 with no uncertainty Taking gamble and possibly receiving nothing
Risk Averse • Lebih memilih sesuatu yang lebih pasti, dan meminimalkan resiko • Contoh: investor memilih return lbh tinggi utk resiko yang sama atau investor lebih memilih resiko yg lbh rendah utk return yang sama
Risk Aversion, neutral and loving • Risk neutral: indifferent between the bet and a certain $50 payment
Risk Aversion, neutral and loving • Risk loving (or risk seeking/loss aversion) Guaranteed payment must be more than $50 to induce him to take the guaranteed option, rather than taking gamble and possibly winning $100
Loss aversion Kondisidimana investor tidakinginmengalamikerugiansehingga investor tersebutbertindaksecaratidakrasional. Merekalebihmemperhatikankondisiuntungataurugi dan tidakmemperhatikanbesarkecilnyanilaiuntungataupunrugi
Contoh tindakan Loss Averse • Mengambilinvestasi yang berisikotinggidenganharapanuntukmengembalikankerugian yang sudahterjadi. • Investor lebih memilih pilihan dg expected return adalah -25% daripada -20% Return -50% dg p=0,5 atau tdk rugi apapun dg p=0,5 Return -20%
Utility theory uncertainty Decision making problem Expected value principle Inadequacy of expected value principle Open and read Bowers, et al (1997) on p. 3-4
Utility theory • Utility is often assumed to be a function of profit or final portofolio wealth, with positive first derivative. • ` • 1. risk seeking • 2. risk neutral • 3.risk averse
Utility function • w : wealth • u(w):utility function • Linear transformation: u*(w) = au(w) + b, a> 0
Utility function • Suppose you face a loss of 20.000 with prob 0,5 and no loss with prob 0,5. • What is the max amount G to pay for complete insurance protection? u(20.000-G)=0,5u(20.000)+0,5u(0) =0,5(0) + 0,5(-1) =-0,5
Utility function • Utility function can be used to compare economic prospects (X and Y). • If decision maker has wealth w, he will select X if E[u(w+X)]>E[u(w+Y)] and indifferent if E[u(w+X)]=E[u(w+Y)]
Insurance and utility • Pure or net premium: price for full insurance coverage, i.e expected loss E(X)= • Loaded premium: H = (1+a) +c a>0, c>0
Insurance and utility expected utility of not buying insurance, current wealth is w U(w-G) = E[u(w-X)] Expected utility of paying G for complete financial protection
Insurance and utility If owner have u(w) = bw + d with b > 0, owner prefers or indiferrents to the insurance if u(w-G) = b(w-G)+d ≥E[u(w-X)]=E[b(w-X)+d] b(w-G)+d ≥b(w- )+d G ≤
Jensen’s inequalities For a random variable X, function u(w) if u”(w)<0 then E[u(X)] ≤ u(E[X]) if u”(w)>0 then E[u(X)] ≥ u(E[X]) Proof: see Bowers, et al (1997) on p. 9
Jensen’s inequality Applying Jensen’s inequality to U(w-G) = E[u(w-X)] We have u(w-G) = E[u(w-X)] ≤ u(w-) because u’(w)>0,u(w) is increasing function w-G ≤ w- G ≥
Utility function for the insurer H: acceptable premium; X: random loss u(w)=E[u(w+H-X)] Jensen’s inequality u(w)=E[u(w+H-X)] ≤ u(w+H-) We can conclude that H ≥
Exponential utility function Check on page 10-11 u(w) = - exp(-w) for all w and a fixed > 0 G doesn’t depend on w
Contoh 1 • Open and read example 1.3.1 p.11 Diketahui : u(w)= -(exp(-5w)), dan X,Y adlh hsl yg mungkin di capai ? X N(5,2) Y N(6,2.5)
Contoh 1 Solution E[u(X)]=-1 > E[u(Y)] = - (exp(1,25)) Dengan demikian, Distribusi X lebih disukai daripada distribusi Y
Family of fractional power utility functions • u(w) = w risk averse cek u’(w) dan u’’(w)
Contoh 2 • Misal • w= 10 dan kehilangan random X UNIF (0,10) Max amount will pay for complete insurance againts the random loss? diperoleh G = 5.5556 , G > E [X]
Family of Quadratic Utility Function u(w) = w - w2 , w < (2)-1 and >0 cek u’(w) dan u’’(w)
Contoh 3 • Diketahui: u(w) = w - w2 , w< 50 retain wealth w with prop p and financial loss with prop (1-p) Sehingga u(w-G) = pu(w) + (1-p)u(w-c)
Contoh 4 Diketahui: Prop properti tdk ada kerugian : 0,75 • Pdf kerugian f(x) = 0,25(0,01exp(-0,01x)), x>0 u(w) = -exp(-0,005w) Hitung: E[X] dan G Cek Ex 1.3.4
Contoh 5 • Diskusikan contoh 1.3.5