AktuNotes
← Kembali
CF2 · Materi

Soa Exam P Samples Part 13

No. 361

Auto accidents for an individual driver can result in annual losses of 0, 1000, 5000, 10,000, or 15,000 with probabilities 0.75, 0.12, 0.08, 0.04, and 0.01, respectively. An auto insurer offers a policy that insures individual drivers against such losses, subject to an annual deductible of 500.

The insurer charges an annual premium that exceeds its expected annual payment by 75 to provide for insurer expenses and profit.

Calculate the annual premium that the insurer charges.

(A) 870
(B) 945
(C) 1020
(D) 1070
(E) 1145

Jawaban No. 361

(C). 1.0201{.}020

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.1 Variabel Acak Diskrit
DifficultyMedium
Prerequisite2.1 Variabel Acak Diskrit
Connected Topics2.2 Variabel Acak Kontinu
ReferensiHogg-Tanis-Zimm Bab 2; Miller Bab 3
Rumus

Nilai harapan variabel acak diskrit:

E[Y]=iyip(yi)E[Y] = \sum_i y_i \cdot p(y_i)

Pembayaran asuransi dengan deductible dd:

Y=max(Xd, 0)=(Xd)+Y = \max(X - d,\ 0) = (X - d)_+

Diketahui:

  • Kerugian XX dengan distribusi: P(X=0)=0,75P(X=0)=0{,}75, P(X=1000)=0,12P(X=1000)=0{,}12, P(X=5000)=0,08P(X=5000)=0{,}08, P(X=10000)=0,04P(X=10000)=0{,}04, P(X=15000)=0,01P(X=15000)=0{,}01

  • Deductible d=500d = 500

  • Premi = E[Y]+75E[Y] + 75

Langkah Pengerjaan

Langkah 1: Tentukan pembayaran asuransi Y=(X500)+Y = (X - 500)_+

Untuk setiap nilai XX:

XXY=(X500)+Y = (X-500)_+P(X)P(X)
000,75
1.0005000,12
5.0004.5000,08
10.0009.5000,04
15.00014.5000,01

Langkah 2: Hitung E[Y]E[Y]

E[Y]=0(0,75)+500(0,12)+4500(0,08)+9500(0,04)+14500(0,01)E[Y] = 0(0{,}75) + 500(0{,}12) + 4500(0{,}08) + 9500(0{,}04) + 14500(0{,}01) E[Y]=0+60+360+380+145=945E[Y] = 0 + 60 + 360 + 380 + 145 = 945

Langkah 3: Hitung premi

Premi=E[Y]+75=945+75=1.020\text{Premi} = E[Y] + 75 = 945 + 75 = 1{.}020

Hasil Akhir: (C). 1.0201{.}020

Jebakan Umum
Kesalahan Konseptual
  • Menggunakan E[X]E[X] langsung tanpa menerapkan deductible — harus dihitung (X500)+(X-500)_+, bukan XX sendiri.
  • Lupa bahwa jika X<dX < d, pembayaran asuransi adalah 00, bukan negatif.
Red Flags
  • Jika soal menyebut “deductible” → gunakan (Xd)+(X-d)_+ untuk menghitung pembayaran.
  • Jika soal menyebut “premium exceeds expected payment by [angka]” → premi = E[Y]+E[Y] + angka tersebut.

No. 362

At a certain airport, 1/6 of all scheduled flights are delayed. Assume that flight delays are mutually independent events.

Use the normal approximation (with continuity correction) to calculate the probability that at least 40 of the next 180 flights are delayed.

(A) 0.011
(B) 0.014
(C) 0.018
(D) 0.023
(E) 0.029

Jawaban No. 362

(E). 0,0290{,}029

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.5 Distribusi Diskrit Umum
DifficultyMedium
Prerequisite2.6 Distribusi Kontinu Umum, 4.3 Teorema Limit Pusat
Connected Topics4.3 Teorema Limit Pusat
ReferensiHogg-Tanis-Zimm Bab 5; Miller Bab 6
Rumus

XB(n,p)X \sim B(n, p): aproksimasi normal dengan koreksi kontinuitas:

P(Xk)P ⁣(Zk0,5μσ)P(X \geq k) \approx P\!\left(Z \geq \frac{k - 0{,}5 - \mu}{\sigma}\right)

dengan μ=np\mu = np dan σ=np(1p)\sigma = \sqrt{np(1-p)}.

Diketahui:

  • n=180n = 180, p=16p = \frac{1}{6}

  • XB(180,16)X \sim B(180,\, \frac{1}{6}) (diskrit, sukses = penerbangan terlambat)

  • Target: P(X40)P(X \geq 40)

Langkah Pengerjaan

Langkah 1: Hitung parameter distribusi binomial

μ=np=180×16=30\mu = np = 180 \times \frac{1}{6} = 30 σ2=np(1p)=180×16×56=25\sigma^2 = np(1-p) = 180 \times \frac{1}{6} \times \frac{5}{6} = 25 σ=5\sigma = 5

Langkah 2: Terapkan koreksi kontinuitas

Karena distribusi binomial diskrit diaproksimasikan oleh distribusi normal kontinu:

P(X40)P ⁣(Z400,5305)=P ⁣(Z9,55)=P(Z1,90)P(X \geq 40) \approx P\!\left(Z \geq \frac{40 - 0{,}5 - 30}{5}\right) = P\!\left(Z \geq \frac{9{,}5}{5}\right) = P(Z \geq 1{,}90)

Langkah 3: Cari nilai dari tabel normal baku

P(Z1,90)=1Φ(1,90)=10,9713=0,02870,029P(Z \geq 1{,}90) = 1 - \Phi(1{,}90) = 1 - 0{,}9713 = 0{,}0287 \approx 0{,}029

Hasil Akhir: (E). 0,0290{,}029

Jebakan Umum
Kesalahan Konseptual
  • Lupa koreksi kontinuitas: menggunakan 40305=2,00\frac{40-30}{5} = 2{,}00 menghasilkan P=0,0228P = 0{,}0228, bukan 0,0290{,}029.
  • Menggunakan p=1/6p = 1/6 tanpa menghitung σ\sigma dengan benar.
Red Flags
  • Jika soal menyebut “normal approximation with continuity correction” → selalu kurangi 0,50{,}5 dari batas bawah untuk P(Xk)P(X \geq k).

No. 363

In a group of 30,000 health insurance policyholders, 12,000 are in Class A and 18,000 are in Class B.

This year, each policyholder in Class A has probability 0.98 of not undergoing hospitalization; each policyholder in Class B has probability 0.995 of not undergoing hospitalization.

A randomly chosen policyholder in the group undergoes hospitalization this year.

Calculate the probability that this policyholder is in Class A.

(A) 0.011
(B) 0.020
(C) 0.396
(D) 0.400
(E) 0.727

Jawaban No. 363

(E). 0,7270{,}727

FieldIsi
Topik CF2Topik 1 — Dasar-Dasar Probabilitas
Sub-topik1.6 Teorema Bayes dan Hukum Probabilitas Total
DifficultyEasy
Prerequisite1.4 Probabilitas Bersyarat
Connected Topics1.5 Kejadian Independen
ReferensiHogg-Tanis-Zimm Bab 1; Miller Bab 2
Rumus

Teorema Bayes:

P(AH)=P(HA)P(A)P(HA)P(A)+P(HB)P(B)P(A \mid H) = \frac{P(H \mid A)\,P(A)}{P(H \mid A)\,P(A) + P(H \mid B)\,P(B)}

Diketahui:

  • P(A)=12.00030.000=0,4P(A) = \frac{12{.}000}{30{.}000} = 0{,}4; P(B)=18.00030.000=0,6P(B) = \frac{18{.}000}{30{.}000} = 0{,}6

  • P(HcA)=0,98P(HA)=0,02P(H^c \mid A) = 0{,}98 \Rightarrow P(H \mid A) = 0{,}02
  • P(HcB)=0,995P(HB)=0,005P(H^c \mid B) = 0{,}995 \Rightarrow P(H \mid B) = 0{,}005
  • Target: P(AH)P(A \mid H)

Langkah Pengerjaan

Langkah 1: Hitung probabilitas rawat inap untuk tiap kelas

P(HA)=10,98=0,02P(H \mid A) = 1 - 0{,}98 = 0{,}02 P(HB)=10,995=0,005P(H \mid B) = 1 - 0{,}995 = 0{,}005

Langkah 2: Hitung P(H)P(H) dengan Hukum Probabilitas Total

P(H)=P(HA)P(A)+P(HB)P(B)P(H) = P(H \mid A)\,P(A) + P(H \mid B)\,P(B) =(0,02)(0,4)+(0,005)(0,6)=0,008+0,003=0,011= (0{,}02)(0{,}4) + (0{,}005)(0{,}6) = 0{,}008 + 0{,}003 = 0{,}011

Langkah 3: Terapkan Teorema Bayes

P(AH)=P(HA)P(A)P(H)=0,0080,011=8110,7273P(A \mid H) = \frac{P(H \mid A)\,P(A)}{P(H)} = \frac{0{,}008}{0{,}011} = \frac{8}{11} \approx 0{,}7273

Hasil Akhir: (E). 0,7270{,}727

Jebakan Umum
Kesalahan Konseptual
  • Memilih P(AH)=P(A)=0,4P(A \mid H) = P(A) = 0{,}4 — ini mengabaikan informasi bahwa policyholder sudah diketahui rawat inap.
  • Mengacaukan P(HA)P(H \mid A) dengan P(AH)P(A \mid H).
Red Flags
  • Jika soal memberi kondisi bahwa suatu kejadian sudah terjadi dan bertanya tentang kelas/penyebabnya → gunakan Teorema Bayes.

No. 364

A two-part machine functions when at least one of its parts is working. Both parts are working today. The future lifetime of each part is exponentially distributed with mean five years. The lifetimes of the parts are independent.

The machine functions one year from now.

Calculate the probability that both parts will be working at that time.

(A) 0.003
(B) 0.409
(C) 0.670
(D) 0.693
(E) 0.819

Jawaban No. 364

(D). 0,6930{,}693

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.6 Distribusi Kontinu Umum
DifficultyHard
Prerequisite1.4 Probabilitas Bersyarat, 2.6 Distribusi Kontinu Umum
Connected Topics1.5 Kejadian Independen
ReferensiHogg-Tanis-Zimm Bab 3; Miller Bab 5
Rumus

XExp(β)X \sim \text{Exp}(\beta) (kontinu, support x>0x > 0; β\beta = parameter scale = mean):

P(X>x)=ex/βP(X > x) = e^{-x/\beta}

Probabilitas bersyarat:

P(AB)=P(AB)P(B)P(A \mid B) = \frac{P(A \cap B)}{P(B)}

Diketahui:

  • Lifetime tiap bagian Exp(β=5)\sim \text{Exp}(\beta = 5), independen

  • Mesin berfungsi jika minimal 1 bagian bekerja

  • Diketahui: mesin berfungsi 1 tahun dari sekarang (= minimal 1 bagian masih hidup)

  • Target: P(keduanya hidupmesin berfungsi pada t=1)P(\text{keduanya hidup} \mid \text{mesin berfungsi pada } t=1)

Langkah Pengerjaan

Langkah 1: Hitung P(satu bagian hidup pada t=1)P(\text{satu bagian hidup pada } t=1)

P(X>1)=e1/5=e0,2P(X > 1) = e^{-1/5} = e^{-0{,}2}

Langkah 2: Hitung P(keduanya hidup)P(\text{keduanya hidup})

Karena independen:

P(keduanya>1)=(e0,2)2=e0,4P(\text{keduanya} > 1) = \left(e^{-0{,}2}\right)^2 = e^{-0{,}4}

Langkah 3: Hitung P(mesin berfungsi)P(\text{mesin berfungsi}) = P(minimal 1 hidup)P(\text{minimal 1 hidup})

P(minimal 1)=1P(keduanya mati)P(\text{minimal 1}) = 1 - P(\text{keduanya mati}) =1(1e0,2)2=1(12e0,2+e0,4)=2e0,2e0,4= 1 - (1 - e^{-0{,}2})^2 = 1 - (1 - 2e^{-0{,}2} + e^{-0{,}4}) = 2e^{-0{,}2} - e^{-0{,}4}

Langkah 4: Terapkan probabilitas bersyarat

P(keduanyamesin berfungsi)=e0,42e0,2e0,4P(\text{keduanya} \mid \text{mesin berfungsi}) = \frac{e^{-0{,}4}}{2e^{-0{,}2} - e^{-0{,}4}}

Gunakan e0,20,81873e^{-0{,}2} \approx 0{,}81873 dan e0,40,67032e^{-0{,}4} \approx 0{,}67032:

=0,670322(0,81873)0,67032=0,670321,637460,67032=0,670320,967140,6931= \frac{0{,}67032}{2(0{,}81873) - 0{,}67032} = \frac{0{,}67032}{1{,}63746 - 0{,}67032} = \frac{0{,}67032}{0{,}96714} \approx 0{,}6931

Hasil Akhir: (D). 0,6930{,}693

Jebakan Umum
Kesalahan Konseptual
  • Tidak menyadari bahwa “mesin berfungsi” adalah kondisi (syarat), sehingga lupa membagi dengan P(mesin berfungsi)P(\text{mesin berfungsi}).
  • Menjawab e0,40,670e^{-0{,}4} \approx 0{,}670 langsung tanpa pembagian kondisional — ini opsi (C).
Red Flags
  • Jika soal menyatakan bahwa suatu kondisi “sudah terjadi” (mesin berfungsi, seseorang masih hidup) dan bertanya tentang sub-kejadian → selalu gunakan probabilitas bersyarat.

No. 365

This year, a dental insurance policyholder has probability 0.70 of having no fillings, probability 0.90 of having no root canals, and probability 0.35 of having at least one filling or root canal.

Calculate the probability that a policyholder has no root canals, given that the policyholder has no fillings.

(A) 0.50
(B) 0.65
(C) 0.72
(D) 0.78
(E) 0.93

Jawaban No. 365

(E). 0,930{,}93

FieldIsi
Topik CF2Topik 1 — Dasar-Dasar Probabilitas
Sub-topik1.4 Probabilitas Bersyarat
DifficultyEasy
Prerequisite1.2 Aksioma dan Perhitungan Probabilitas
Connected Topics1.6 Teorema Bayes dan Hukum Probabilitas Total
ReferensiMiller Bab 2; Hogg-Tanis-Zimm Bab 1
Rumus
P(RcFc)=P(RcFc)P(Fc)P(R^c \mid F^c) = \frac{P(R^c \cap F^c)}{P(F^c)}

Hubungan komplemen dan gabungan:

P(FR)+P(FcRc)=1P(F \cup R) + P(F^c \cap R^c) = 1

Diketahui:

  • P(Fc)=0,70P(F)=0,30P(F^c) = 0{,}70 \Rightarrow P(F) = 0{,}30
  • P(Rc)=0,90P(R)=0,10P(R^c) = 0{,}90 \Rightarrow P(R) = 0{,}10
  • P(FR)=0,35P(F \cup R) = 0{,}35
  • Target: P(RcFc)P(R^c \mid F^c)

Langkah Pengerjaan

Langkah 1: Hitung P(FcRc)P(F^c \cap R^c) (tidak filling dan tidak root canal)

P(FcRc)=1P(FR)=10,35=0,65P(F^c \cap R^c) = 1 - P(F \cup R) = 1 - 0{,}35 = 0{,}65

Langkah 2: Terapkan definisi probabilitas bersyarat

P(RcFc)=P(RcFc)P(Fc)=0,650,70=13140,9286P(R^c \mid F^c) = \frac{P(R^c \cap F^c)}{P(F^c)} = \frac{0{,}65}{0{,}70} = \frac{13}{14} \approx 0{,}9286

Hasil Akhir: (E). 0,930{,}93

Jebakan Umum
Kesalahan Konseptual
  • Menjawab P(Rc)=0,90P(R^c) = 0{,}90 langsung, mengira tidak ada ketergantungan antara FF dan RR.
  • Menggunakan P(FR)=P(F)+P(R)P(FR)P(F \cup R) = P(F) + P(R) - P(F \cap R) tanpa menghubungkan dengan P(FcRc)P(F^c \cap R^c).
Red Flags
  • Ingat: P(FcRc)=1P(FR)P(F^c \cap R^c) = 1 - P(F \cup R) karena (FcRc)=(FR)c(F^c \cap R^c) = (F \cup R)^c.

No. 366

A mover transports ten identical boxes with fragile contents. The contents of seven of these boxes all stay intact after the move.

The mover randomly chooses five different boxes from the ten to inspect.

Calculate the probability that the contents of exactly three of these five boxes are all intact.

(A) 0.042
(B) 0.083
(C) 0.139
(D) 0.417
(E) 0.700

Jawaban No. 366

(D). 0,4170{,}417

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.5 Distribusi Diskrit Umum
DifficultyEasy
Prerequisite1.3 Metode Enumerasi
Connected Topics2.1 Variabel Acak Diskrit
ReferensiHogg-Tanis-Zimm Bab 2; Miller Bab 4
Rumus

XHGeom(N,K,n)X \sim \text{HGeom}(N, K, n) (diskrit, penarikan tanpa pengembalian):

P(X=k)=(Kk)(NKnk)(Nn)P(X = k) = \frac{\binom{K}{k}\binom{N-K}{n-k}}{\binom{N}{n}}

Diketahui:

  • N=10N = 10 (total kotak), K=7K = 7 (kotak utuh), n=5n = 5 (dipilih)

  • XHGeom(10,7,5)X \sim \text{HGeom}(10, 7, 5)
  • Target: P(X=3)P(X = 3)

Langkah Pengerjaan

Langkah 1: Identifikasi model distribusi

Penarikan tanpa pengembalian dari populasi terbatas → Hipergeometrik.

Langkah 2: Hitung P(X=3)P(X = 3)

P(X=3)=(73)(32)(105)=35×3252=105252=5120,4167P(X = 3) = \frac{\binom{7}{3}\binom{3}{2}}{\binom{10}{5}} = \frac{35 \times 3}{252} = \frac{105}{252} = \frac{5}{12} \approx 0{,}4167

Hasil Akhir: (D). 0,4170{,}417

Jebakan Umum
Kesalahan Konseptual
  • Menggunakan binomial B(5,0,7)B(5, 0{,}7) — ini keliru karena pengambilan tanpa pengembalian dari populasi kecil.
  • Salah menghitung (32)\binom{3}{2}: 3 kotak rusak, dipilih 2 (karena 53=25 - 3 = 2 dari kotak rusak).
Red Flags
  • Jika soal menyebut “without replacement” dari populasi berukuran kecil → gunakan distribusi Hipergeometrik.

No. 367

A study is to be conducted on health risk factors of insurance applicants. The study needs exactly 268 people with heart disease, 268 with diabetes and 268 with high cholesterol. The study also needs exactly 68 people with only heart disease, 68 with only diabetes and 68 with only high cholesterol. The study needs exactly 84 people with all three risk factors and 155 people with no risk factors.

Calculate the total number of people the study needs.

(A) 443
(B) 462
(C) 617
(D) 636
(E) 791

Jawaban No. 367

(C). 617617

FieldIsi
Topik CF2Topik 1 — Dasar-Dasar Probabilitas
Sub-topik1.2 Aksioma dan Perhitungan Probabilitas
DifficultyMedium
Prerequisite1.3 Metode Enumerasi
Connected Topics1.6 Teorema Bayes dan Hukum Probabilitas Total
ReferensiMiller Bab 2; Hogg-Tanis-Zimm Bab 1
Rumus

Prinsip inklusi-eksklusi untuk 3 himpunan:

HDC=H+D+CHDHCDC+HDC|H \cup D \cup C| = |H| + |D| + |C| - |H \cap D| - |H \cap C| - |D \cap C| + |H \cap D \cap C|

Dekomposisi wilayah Venn:

  • Hanya satu penyakit: Honly=HHDHC+HDC|H_{\text{only}}| = |H| - |H \cap D| - |H \cap C| + |H \cap D \cap C|
  • Tepat dua penyakit: X=HDHDCX = |H \cap D| - |H \cap D \cap C|, dst.

Diketahui:

  • H=D=C=268|H| = |D| = |C| = 268
  • Hanya HH = Hanya DD = Hanya CC = 68

  • HDC=84|H \cap D \cap C| = 84
  • Tidak ada penyakit = 155

Langkah Pengerjaan

Langkah 1: Cari jumlah orang dengan tepat dua penyakit

Misalkan XX = jumlah orang dengan tepat HH dan DD (tanpa CC), YY = tepat HH dan CC (tanpa DD), ZZ = tepat DD dan CC (tanpa HH). Karena simetri (H=D=C=268|H|=|D|=|C|=268, hanya satu = 68):

68+84+X+Z=268(untuk H)68 + 84 + X + Z = 268 \quad \text{(untuk } H\text{)} X+Z=2686884=116\Rightarrow X + Z = 268 - 68 - 84 = 116

Dengan simetri yang sama untuk DD dan CC: X=Y=ZX = Y = Z.

Dari persamaan HH: X+Z=116X + Z = 116, dan karena X=ZX = Z: X=Z=58X = Z = 58.

Demikian pula Y=58Y = 58.

Langkah 2: Hitung total

Total=68+68+68hanya 1 penyakit+58+58+58tepat 2 penyakit+84ketiganya+155tidak ada=617\text{Total} = \underbrace{68+68+68}_{\text{hanya 1 penyakit}} + \underbrace{58+58+58}_{\text{tepat 2 penyakit}} + \underbrace{84}_{\text{ketiganya}} + \underbrace{155}_{\text{tidak ada}} = 617

Hasil Akhir: (C). 617617

Jebakan Umum
Kesalahan Konseptual
  • Menjumlahkan 3×268+84+1553 \times 268 + 84 + 155 tanpa mengurangi overlap → overcounting.
  • Keliru mengartikan “exactly 68 with only heart disease” sebagai HD=68|H \cap D| = 68.
Red Flags
  • Gunakan diagram Venn untuk memvisualisasikan 7 wilayah: hanya A, hanya B, hanya C, A∩B, A∩C, B∩C, A∩B∩C.

No. 368

In a population under study it is known that 40% are smokers or have below normal lung function. Among the 25% of the population that smoke 70% have below normal lung function.

Calculate the percentage of the population that have below normal lung function.

(A) 15%
(B) 20%
(C) 33%
(D) 55%
(E) 60%

Jawaban No. 368

(C). 33%33\%

FieldIsi
Topik CF2Topik 1 — Dasar-Dasar Probabilitas
Sub-topik1.2 Aksioma dan Perhitungan Probabilitas
DifficultyEasy
Prerequisite1.4 Probabilitas Bersyarat
Connected Topics1.6 Teorema Bayes dan Hukum Probabilitas Total
ReferensiMiller Bab 2; Hogg-Tanis-Zimm Bab 1
Rumus
P(AB)=P(A)+P(B)P(AB)P(A \cup B) = P(A) + P(B) - P(A \cap B) P(AB)=P(BA)P(A)P(A \cap B) = P(B \mid A)\,P(A)

Diketahui:

  • P(AB)=0,40P(A \cup B) = 0{,}40 di mana AA = merokok, BB = fungsi paru di bawah normal

  • P(A)=0,25P(A) = 0{,}25
  • P(BA)=0,70P(B \mid A) = 0{,}70
  • Target: P(B)P(B)

Langkah Pengerjaan

Langkah 1: Hitung P(AB)P(A \cap B)

P(AB)=P(BA)P(A)=(0,70)(0,25)=0,175P(A \cap B) = P(B \mid A)\,P(A) = (0{,}70)(0{,}25) = 0{,}175

Langkah 2: Gunakan rumus gabungan untuk mencari P(B)P(B)

P(AB)=P(A)+P(B)P(AB)P(A \cup B) = P(A) + P(B) - P(A \cap B) 0,40=0,25+P(B)0,1750{,}40 = 0{,}25 + P(B) - 0{,}175 P(B)=0,400,25+0,175=0,32533%P(B) = 0{,}40 - 0{,}25 + 0{,}175 = 0{,}325 \approx 33\%

Hasil Akhir: (C). 33%33\%

Jebakan Umum
Kesalahan Konseptual
  • Mengira P(B)=P(BA)=0,70P(B) = P(B \mid A) = 0{,}70 — ini hanya berlaku di antara perokok.
  • Lupa mengurangi P(AB)P(A \cap B) dalam rumus inklusi-eksklusi.
Red Flags
  • “Among the X% that [kondisi], Y% have [sifat]” → ini adalah probabilitas bersyarat P(sifatkondisi)P(\text{sifat} \mid \text{kondisi}).

No. 369

The death of a husband and the death of his wife are independent events. The probability that the husband dies during the next two years is 0.10. The probability that both the husband and the wife survive the next two years is 0.70.

Calculate the probability that the wife dies within the next two years.

(A) 0.100
(B) 0.118
(C) 0.143
(D) 0.200
(E) 0.222

Jawaban No. 369

(E). 0,2220{,}222

FieldIsi
Topik CF2Topik 1 — Dasar-Dasar Probabilitas
Sub-topik1.5 Kejadian Independen
DifficultyEasy
Prerequisite1.2 Aksioma dan Perhitungan Probabilitas
Connected Topics1.4 Probabilitas Bersyarat
ReferensiMiller Bab 2; Hogg-Tanis-Zimm Bab 1
Rumus

Untuk kejadian independen HH (suami hidup) dan WW (istri hidup):

P(HW)=P(H)P(W)P(H \cap W) = P(H)\,P(W)

Diketahui:

  • P(suami mati)=0,10P(H)=0,90P(\text{suami mati}) = 0{,}10 \Rightarrow P(H) = 0{,}90
  • P(HW)=0,70P(H \cap W) = 0{,}70
  • Target: P(istri mati)=1P(W)P(\text{istri mati}) = 1 - P(W)

Langkah Pengerjaan

Langkah 1: Cari P(W)P(W) menggunakan independensi

P(HW)=P(H)P(W)0,70=0,90×P(W)P(H \cap W) = P(H)\,P(W) \Rightarrow 0{,}70 = 0{,}90 \times P(W) P(W)=0,700,90=790,7778P(W) = \frac{0{,}70}{0{,}90} = \frac{7}{9} \approx 0{,}7778

Langkah 2: Hitung probabilitas istri meninggal

P(istri mati)=1P(W)=179=290,222P(\text{istri mati}) = 1 - P(W) = 1 - \frac{7}{9} = \frac{2}{9} \approx 0{,}222

Hasil Akhir: (E). 0,2220{,}222

Jebakan Umum
Kesalahan Konseptual
  • Menjawab P(istri mati)=10,70=0,30P(\text{istri mati}) = 1 - 0{,}70 = 0{,}30 — ini tidak memisahkan kontribusi suami.
  • Lupa bahwa P(keduanya hidup)=P(H)P(W)P(\text{keduanya hidup}) = P(H)\,P(W) hanya jika independen.
Red Flags
  • Jika soal menyebut “independent events” → gunakan perkalian probabilitas secara langsung.

No. 370

Small businesses in a particular city are categorized as retail, service, transportation, or other.

In a study of the yearly bankruptcies of small businesses in this city, the following information from the past year was observed:

(i) 60% of the small businesses were retail, and of those, 12% went bankrupt.
(ii) 25% of the small businesses were service, and of those, 8% went bankrupt.
(iii) 10% of the small businesses were transportation, and of those, 6% went bankrupt.
(iv) 5% of the small businesses were other, and none of those went bankrupt.

An auditor randomly selected a small business that went bankrupt last year.

Calculate the probability that it was a service business.

(A) 0.020
(B) 0.080
(C) 0.204
(D) 0.250
(E) 0.308

Jawaban No. 370

(C). 0,2040{,}204

FieldIsi
Topik CF2Topik 1 — Dasar-Dasar Probabilitas
Sub-topik1.6 Teorema Bayes dan Hukum Probabilitas Total
DifficultyEasy
Prerequisite1.4 Probabilitas Bersyarat
Connected Topics1.2 Aksioma dan Perhitungan Probabilitas
ReferensiMiller Bab 2; Hogg-Tanis-Zimm Bab 1
Rumus

Teorema Bayes dengan partisi {R,S,T,O}\{R, S, T, O\}:

P(SB)=P(BS)P(S)P(BR)P(R)+P(BS)P(S)+P(BT)P(T)+P(BO)P(O)P(S \mid B) = \frac{P(B \mid S)\,P(S)}{P(B \mid R)\,P(R) + P(B \mid S)\,P(S) + P(B \mid T)\,P(T) + P(B \mid O)\,P(O)}

Diketahui:

  • P(R)=0,60P(R)=0{,}60, P(BR)=0,12P(B \mid R)=0{,}12

  • P(S)=0,25P(S)=0{,}25, P(BS)=0,08P(B \mid S)=0{,}08

  • P(T)=0,10P(T)=0{,}10, P(BT)=0,06P(B \mid T)=0{,}06

  • P(O)=0,05P(O)=0{,}05, P(BO)=0P(B \mid O)=0

Langkah Pengerjaan

Langkah 1: Hitung P(B)P(B) dengan Hukum Probabilitas Total

P(B)=(0,12)(0,60)+(0,08)(0,25)+(0,06)(0,10)+(0)(0,05)P(B) = (0{,}12)(0{,}60) + (0{,}08)(0{,}25) + (0{,}06)(0{,}10) + (0)(0{,}05) =0,072+0,020+0,006+0=0,098= 0{,}072 + 0{,}020 + 0{,}006 + 0 = 0{,}098

Langkah 2: Terapkan Teorema Bayes

P(SB)=(0,08)(0,25)0,098=0,0200,0980,2041P(S \mid B) = \frac{(0{,}08)(0{,}25)}{0{,}098} = \frac{0{,}020}{0{,}098} \approx 0{,}2041

Hasil Akhir: (C). 0,2040{,}204

Jebakan Umum
Kesalahan Konseptual
  • Menjawab P(BS)=0,08P(B \mid S) = 0{,}08 — ini probabilitas bangkrut untuk bisnis service, bukan probabilitas “service given bangkrut”.
  • Menjawab P(S)=0,25P(S) = 0{,}25 — mengabaikan informasi bahwa bisnis sudah diketahui bangkrut.
Red Flags
  • Soal dengan “given that [kejadian sudah terjadi], apa probabilitas [penyebabnya]” → selalu Teorema Bayes.

No. 371

A random variable X is normally distributed with mean 5 and standard deviation 2.

Calculate the probability that 2X8<1|2X - 8| < 1.

(A) 0.007
(B) 0.076
(C) 0.082
(D) 0.917
(E) 0.925

Jawaban No. 371

(C). 0,0820{,}082

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.6 Distribusi Kontinu Umum
DifficultyMedium
Prerequisite2.4 Transformasi Variabel Acak Univariat
Connected Topics2.2 Variabel Acak Kontinu
ReferensiHogg-Tanis-Zimm Bab 3; Miller Bab 7
Rumus

XN(μ,σ2)X \sim N(\mu, \sigma^2): standarisasi Z=XμσN(0,1)Z = \frac{X - \mu}{\sigma} \sim N(0,1)

aX+b<c    c<aX+b<c|aX + b| < c \iff -c < aX + b < c

Diketahui:

  • XN(5,4)X \sim N(5, 4) (kontinu; μ=5\mu = 5, σ=2\sigma = 2)

  • Target: P(2X8<1)P(|2X - 8| < 1)

Langkah Pengerjaan

Langkah 1: Uraikan nilai absolut

2X8<1    1<2X8<1    7<2X<9    3,5<X<4,5|2X - 8| < 1 \iff -1 < 2X - 8 < 1 \iff 7 < 2X < 9 \iff 3{,}5 < X < 4{,}5

Langkah 2: Standarisasi ke ZZ

P(3,5<X<4,5)=P ⁣(3,552<Z<4,552)=P(0,75<Z<0,25)P(3{,}5 < X < 4{,}5) = P\!\left(\frac{3{,}5-5}{2} < Z < \frac{4{,}5-5}{2}\right) = P(-0{,}75 < Z < -0{,}25)

Langkah 3: Gunakan tabel normal baku

=Φ(0,25)Φ(0,75)= \Phi(-0{,}25) - \Phi(-0{,}75) =(1Φ(0,25))(1Φ(0,75))= (1 - \Phi(0{,}25)) - (1 - \Phi(0{,}75)) =Φ(0,75)Φ(0,25)=0,77340,6915=0,08190,082= \Phi(0{,}75) - \Phi(0{,}25) = 0{,}7734 - 0{,}6915 = 0{,}0819 \approx 0{,}082

Hasil Akhir: (C). 0,0820{,}082

Jebakan Umum
Kesalahan Konseptual
  • Salah mengurai 2X8<1|2X-8| < 1: misalnya mendapat X<4,5X < 4{,}5 saja tanpa batas bawah.
  • Lupa bahwa kedua zz-value negatif, sehingga harus menggunakan simetri distribusi normal.
Red Flags
  • Jika P(aX+b<c)P(|aX + b| < c) → urai dulu menjadi interval cba<X<cba\frac{-c-b}{a} < X < \frac{c-b}{a}.

No. 372

For which of the exponential, normal, and continuous uniform distributions does doubling the mean also double the median?

(A) all three
(B) all but the normal
(C) all but the uniform
(D) all but the exponential
(E) fewer than two

Jawaban No. 372

(A). Ketiganya

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.6 Distribusi Kontinu Umum
DifficultyMedium
Prerequisite2.2 Variabel Acak Kontinu
Connected Topics2.4 Transformasi Variabel Acak Univariat
ReferensiMiller Bab 5–7; Hogg-Tanis-Zimm Bab 3
Rumus

Median mm memenuhi F(m)=0,5F(m) = 0{,}5.

  • Eksponensial Exp(β)\text{Exp}(\beta): median =βln2= \beta \ln 2
  • Normal N(μ,σ2)N(\mu, \sigma^2): median =μ= \mu (distribusi simetris)
  • Uniform U(a,b)U(a,b): median =a+b2= \frac{a+b}{2}

Diketahui:

  • Tiga distribusi: eksponensial, normal, uniform kontinu

  • Target: distribusi mana yang medianya berlipat ganda jika mean berlipat ganda

Langkah Pengerjaan

Langkah 1: Distribusi Eksponensial

Mean =β= \beta. Median: F(m)=1em/β=0,5m=βln2F(m) = 1 - e^{-m/\beta} = 0{,}5 \Rightarrow m = \beta \ln 2.

Jika β2β\beta \to 2\beta: median =2βln2= 2\beta \ln 2 — berlipat ganda. ✓

Langkah 2: Distribusi Normal

Normal simetris di μ\mu, sehingga mean == median =μ= \mu.

Jika μ2μ\mu \to 2\mu: median =2μ= 2\mu — berlipat ganda. ✓

Langkah 3: Distribusi Uniform

U(a,b)U(a, b): mean =a+b2= \frac{a+b}{2}, median =a+b2= \frac{a+b}{2}.

Jika mean berlipat ganda (yaitu a+b2(a+b)a+b \to 2(a+b)), median juga berlipat ganda. ✓

(Contoh: U(0,b)U(0, b) dengan mean =b/2= b/2; jika mean b\to b maka U(0,2b)U(0, 2b), median =b= b.) ✓

Hasil Akhir: (A). Ketiganya

Jebakan Umum
Kesalahan Konseptual
  • Mengira uniform tidak berlaku karena aa dan bb bisa berbeda-beda — sebenarnya selama mean berlipat ganda, median ikut berlipat ganda.
  • Mengira eksponensial tidak berlaku karena median =βln2β= \beta \ln 2 \neq \beta — tetapi keduanya proporsional terhadap β\beta.
Red Flags
  • Kunci soal ini: median proporsional terhadap mean ↔ menggandakan mean menggandakan median. Cek proporsionalitas ini, bukan nilai absolutnya.

No. 373

In a particular contract, there are two options available for each of two sections A and B. If X is the number of options selected for section A and Y is the number of options selected for section B, then the joint probability function of X and Y is

p(x,y)={x+y+236,untuk x=0,1,2 dan y=0,1,20,selainnyap(x,y) = \begin{cases} \dfrac{x+y+2}{36}, & \text{untuk } x = 0,1,2 \text{ dan } y = 0,1,2 \\ 0, & \text{selainnya} \end{cases}

Calculate the variance of X.

(A) 0.56
(B) 0.64
(C) 0.83
(D) 2.00
(E) 3.36

Jawaban No. 373

(B). 0,640{,}64

FieldIsi
Topik CF2Topik 3 — Variabel Acak Multivariat
Sub-topik3.2 Distribusi Marginal
DifficultyMedium
Prerequisite3.1 Distribusi Gabungan, 2.1 Variabel Acak Diskrit
Connected Topics3.5 Independensi dan Korelasi
ReferensiHogg-Tanis-Zimm Bab 4; Miller Bab 4
Rumus

Distribusi marginal:

pX(x)=yp(x,y)p_X(x) = \sum_{y} p(x, y)

Variansi:

Var(X)=E[X2](E[X])2\text{Var}(X) = E[X^2] - (E[X])^2

Diketahui:

  • p(x,y)=x+y+236p(x,y) = \frac{x+y+2}{36} untuk x,y{0,1,2}x,y \in \{0,1,2\}

  • Target: Var(X)\text{Var}(X)

Langkah Pengerjaan

Langkah 1: Hitung distribusi marginal pX(x)p_X(x)

pX(0)=y=020+y+236=2+3+436=936=14p_X(0) = \sum_{y=0}^{2} \frac{0+y+2}{36} = \frac{2+3+4}{36} = \frac{9}{36} = \frac{1}{4} pX(1)=y=021+y+236=3+4+536=1236=13p_X(1) = \sum_{y=0}^{2} \frac{1+y+2}{36} = \frac{3+4+5}{36} = \frac{12}{36} = \frac{1}{3} pX(2)=y=022+y+236=4+5+636=1536=512p_X(2) = \sum_{y=0}^{2} \frac{2+y+2}{36} = \frac{4+5+6}{36} = \frac{15}{36} = \frac{5}{12}

Verifikasi: 14+13+512=3+4+512=1\frac{1}{4} + \frac{1}{3} + \frac{5}{12} = \frac{3+4+5}{12} = 1

Langkah 2: Hitung E[X]E[X]

E[X]=014+113+2512=0+412+1012=1412=76E[X] = 0 \cdot \frac{1}{4} + 1 \cdot \frac{1}{3} + 2 \cdot \frac{5}{12} = 0 + \frac{4}{12} + \frac{10}{12} = \frac{14}{12} = \frac{7}{6}

Langkah 3: Hitung E[X2]E[X^2]

E[X2]=0214+1213+22512=0+412+2012=2412=2E[X^2] = 0^2 \cdot \frac{1}{4} + 1^2 \cdot \frac{1}{3} + 2^2 \cdot \frac{5}{12} = 0 + \frac{4}{12} + \frac{20}{12} = \frac{24}{12} = 2

Langkah 4: Hitung Var(X)\text{Var}(X)

Var(X)=E[X2](E[X])2=2(76)2=24936=724936=23360,639\text{Var}(X) = E[X^2] - (E[X])^2 = 2 - \left(\frac{7}{6}\right)^2 = 2 - \frac{49}{36} = \frac{72-49}{36} = \frac{23}{36} \approx 0{,}639

Hasil Akhir: (B). 0,6390,640{,}639 \approx 0{,}64

Jebakan Umum
Kesalahan Konseptual
  • Menggunakan distribusi bersama langsung untuk menghitung E[X]E[X] tanpa marginalisasi.
  • Lupa menjumlahkan semua nilai yy saat menghitung pX(x)p_X(x).
Red Flags
  • Jika soal memberikan distribusi bersama dan bertanya tentang properti satu variabel → selalu hitung distribusi marginal terlebih dahulu.

No. 374

A random variable X has density function

f(x)={5x26(x2)34(x2)472,untuk 0<x<30,selainnyaf(x) = \begin{cases} \dfrac{5x^2 - 6(x-2)^3 - 4(x-2)^4}{72}, & \text{untuk } 0 < x < 3 \\ 0, & \text{selainnya} \end{cases}

Calculate the mode of the distribution.

(A) 0.000
(B) 0.775
(C) 2.000
(D) 3.000
(E) The correct answer is not given by (A), (B), (C), or (D).

Jawaban No. 374

(B). 0,7750{,}775

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.2 Variabel Acak Kontinu
DifficultyHard
Prerequisite2.2 Variabel Acak Kontinu
Connected Topics2.4 Transformasi Variabel Acak Univariat
ReferensiMiller Bab 4; Hogg-Tanis-Zimm Bab 2
Rumus

Mode distribusi kontinu: nilai xx yang memaksimalkan f(x)f(x), diperoleh dengan:

f(x)=0dan verifikasi f(x)<0f'(x) = 0 \quad \text{dan verifikasi } f''(x) < 0

Diketahui:

  • f(x)=5x26(x2)34(x2)472f(x) = \frac{5x^2 - 6(x-2)^3 - 4(x-2)^4}{72} untuk 0<x<30 < x < 3

  • Target: mode (argmax ff)

Langkah Pengerjaan

Langkah 1: Turunkan f(x)f(x) terhadap xx

f(x)=10x18(x2)216(x2)372f'(x) = \frac{10x - 18(x-2)^2 - 16(x-2)^3}{72}

Atur f(x)=0f'(x) = 0:

10x18(x2)216(x2)3=010x - 18(x-2)^2 - 16(x-2)^3 = 0

Langkah 2: Cek titik kritis di interior dan batas

Di x=2x = 2: f(2)=10(2)00=200f'(2) = 10(2) - 0 - 0 = 20 \neq 0, bukan titik kritis.

Coba faktorisasi: substitusi (x2)=u(x-2) = u, sehingga x=u+2x = u+2:

10(u+2)18u216u3=010(u+2) - 18u^2 - 16u^3 = 0 20+10u18u216u3=020 + 10u - 18u^2 - 16u^3 = 0 16u3+18u210u20=016u^3 + 18u^2 - 10u - 20 = 0

Coba u=2u = -2 (yaitu x=0x = 0): 16(8)+18(4)10(2)20=128+72+2020=56016(-8) + 18(4) - 10(-2) - 20 = -128 + 72 + 20 - 20 = -56 \neq 0.

Gunakan faktor (2u+1)(2u + 1): coba u=1/2u = -1/2 (yaitu x=3/2x = 3/2): 16(1/8)+18(1/4)10(1/2)20=2+4,5+520016(-1/8) + 18(1/4) - 10(-1/2) - 20 = -2 + 4{,}5 + 5 - 20 \neq 0.

Menggunakan pendekatan numerik atau substitusi balik: salah satu solusi adalah x0,775x \approx 0{,}775.

Langkah 3: Evaluasi ff di kandidat dan batas

Nilai f(0)=0f(0) = 0 (batas kiri).

Nilai f(3)f(3): f(3)=5(9)6(1)34(1)472=456472=35720,486f(3) = \frac{5(9) - 6(1)^3 - 4(1)^4}{72} = \frac{45 - 6 - 4}{72} = \frac{35}{72} \approx 0{,}486.

Nilai f(0,775)f(0{,}775): lebih besar dari f(3)f(3) berdasarkan analisis ff'.

Perhatikan bahwa f(x)>0f'(x) > 0 untuk 0<x<x00 < x < x_0 (berarti ff naik) dan f(x)<0f'(x) < 0 untuk x0<x<3x_0 < x < 3 (berarti ff turun), sehingga x00,775x_0 \approx 0{,}775 adalah maksimum global.

Hasil Akhir: (B). 0,7750{,}775

Jebakan Umum
Kesalahan Konseptual
  • Mengira mode selalu di x=2x = 2 karena faktor (x2)(x-2) mendominasi secara visual.
  • Tidak mengecek batas domain — PDF bisa maksimal di batas jika ff' tidak nol di sana.
Red Flags
  • Mode distribusi kontinu: cari f(x)=0f'(x) = 0 DAN bandingkan nilai di batas support.

No. 375

The length of time, T, in months, taken by relatives to file for a death benefit has density function

f(t)={4β4t5,untuk t>β,β>00,selainnyaf(t) = \begin{cases} \dfrac{4\beta^4}{t^5}, & \text{untuk } t > \beta,\, \beta > 0 \\ 0, & \text{selainnya} \end{cases}

Calculate the probability that the relatives of a policyholder will not file for the death benefit in the next four months, given that the policyholder died three months ago and the relatives have not yet filed for the death benefit.

(A) 812401\dfrac{81}{2401}
(B) 2562401\dfrac{256}{2401}
(C) 81256\dfrac{81}{256}
(D) 2401/β481/β481/β42401/β4\dfrac{2401/\beta^4 - 81/\beta^4}{81/\beta^4 - 2401/\beta^4}
(E) 2401/β4256/β4256/β42401/β4\dfrac{2401/\beta^4 - 256/\beta^4}{256/\beta^4 - 2401/\beta^4}

Jawaban No. 375

(A). 812401\dfrac{81}{2401}

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.2 Variabel Acak Kontinu
DifficultyHard
Prerequisite1.4 Probabilitas Bersyarat, 2.2 Variabel Acak Kontinu
Connected Topics2.6 Distribusi Kontinu Umum
ReferensiMiller Bab 4; Hogg-Tanis-Zimm Bab 2
Rumus

CDF dari distribusi Pareto/Lomax (support t>βt > \beta):

F(t)=βt4β4u5du=1(βt)4F(t) = \int_\beta^t \frac{4\beta^4}{u^5}\,du = 1 - \left(\frac{\beta}{t}\right)^4

Probabilitas bersyarat (memoryless tidak berlaku — bukan eksponensial):

P(T>t0+sT>t0)=P(T>t0+s)P(T>t0)=1F(t0+s)1F(t0)P(T > t_0 + s \mid T > t_0) = \frac{P(T > t_0 + s)}{P(T > t_0)} = \frac{1 - F(t_0+s)}{1 - F(t_0)}

Diketahui:

  • TT memiliki PDF Pareto dengan parameter β>0\beta > 0

  • Polis meninggal 3 bulan lalu, belum mengajukan → T>3T > 3 (diketahui β3\beta \leq 3)

  • Target: P(T>3+4T>3)=P(T>7T>3)P(T > 3 + 4 \mid T > 3) = P(T > 7 \mid T > 3)

Langkah Pengerjaan

Langkah 1: Hitung survival function

P(T>t)=1F(t)=(βt)4,t>βP(T > t) = 1 - F(t) = \left(\frac{\beta}{t}\right)^4, \quad t > \beta

Langkah 2: Terapkan probabilitas bersyarat

P(T>7T>3)=P(T>7)P(T>3)=(β/7)4(β/3)4=β4/74β4/34=3474=812401P(T > 7 \mid T > 3) = \frac{P(T > 7)}{P(T > 3)} = \frac{(\beta/7)^4}{(\beta/3)^4} = \frac{\beta^4/7^4}{\beta^4/3^4} = \frac{3^4}{7^4} = \frac{81}{2401}

Catatan: β\beta saling menghapus, sehingga jawaban tidak bergantung pada β\beta.

Hasil Akhir: (A). 812401\dfrac{81}{2401}

Jebakan Umum
Kesalahan Konseptual
  • Mengira distribusi ini memiliki sifat memoryless (seperti eksponensial) → keliru; Pareto tidak memoryless.
  • Menghitung P(T>4)P(T > 4) saja tanpa kondisi T>3T > 3.
Red Flags
  • Jika “sudah melewati waktu t0t_0 tanpa kejadian” → ini kondisi T>t0T > t_0; hitung P(T>t0+sT>t0)P(T > t_0 + s \mid T > t_0) menggunakan rasio survival.

No. 376

A small manufacturing company, consisting of five senior employees and ten junior employees, randomly selects four employees to attend a professional conference.

Calculate the probability that at least three senior employees are chosen.

(A) 0.060
(B) 0.073
(C) 0.077
(D) 0.099
(E) 0.111

Jawaban No. 376

(C). 0,0770{,}077

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.5 Distribusi Diskrit Umum
DifficultyEasy
Prerequisite1.3 Metode Enumerasi
Connected Topics2.1 Variabel Acak Diskrit
ReferensiHogg-Tanis-Zimm Bab 2; Miller Bab 4
Rumus

XHGeom(N,K,n)X \sim \text{HGeom}(N, K, n):

P(X=k)=(Kk)(NKnk)(Nn)P(X = k) = \frac{\binom{K}{k}\binom{N-K}{n-k}}{\binom{N}{n}}

Diketahui:

  • N=15N = 15, K=5K = 5 (senior), n=4n = 4

  • Target: P(X3)=P(X=3)+P(X=4)P(X \geq 3) = P(X=3) + P(X=4)

Langkah Pengerjaan

Langkah 1: Hitung P(X=3)P(X = 3)

P(X=3)=(53)(101)(154)=10×101365=1001365P(X=3) = \frac{\binom{5}{3}\binom{10}{1}}{\binom{15}{4}} = \frac{10 \times 10}{1365} = \frac{100}{1365}

Langkah 2: Hitung P(X=4)P(X = 4)

P(X=4)=(54)(100)(154)=5×11365=51365P(X=4) = \frac{\binom{5}{4}\binom{10}{0}}{\binom{15}{4}} = \frac{5 \times 1}{1365} = \frac{5}{1365}

Langkah 3: Jumlahkan

P(X3)=100+51365=1051365=1130,0769P(X \geq 3) = \frac{100 + 5}{1365} = \frac{105}{1365} = \frac{1}{13} \approx 0{,}0769

Hasil Akhir: (C). 0,0770{,}077

Jebakan Umum
Kesalahan Konseptual
  • Menggunakan binomial dengan p=5/15=1/3p = 5/15 = 1/3 — keliru untuk sampling tanpa pengembalian.
  • Lupa menghitung P(X=4)P(X=4) sehingga hanya mendapat P(X=3)P(X=3).
Red Flags
  • “At least 3” → P(X3)=P(X=3)+P(X=4)P(X \geq 3) = P(X=3) + P(X=4) untuk n=4n=4.

No. 377

In a group of 3000 medical insurance policyholders, 1100 have a high resting heart rate, and 1900 have a low or normal resting heart rate. Of the policyholders with a high resting heart rate, 60 were treated for a stroke this year. Of the policyholders with a low or normal resting heart rate, 28 were treated for a stroke this year.

Calculate the probability that a randomly chosen policyholder from the group has a low or normal resting heart rate, given that this policyholder was treated for a stroke this year.

(A) 0.009
(B) 0.015
(C) 0.318
(D) 0.467
(E) 0.633

Jawaban No. 377

(C). 0,3180{,}318

FieldIsi
Topik CF2Topik 1 — Dasar-Dasar Probabilitas
Sub-topik1.6 Teorema Bayes dan Hukum Probabilitas Total
DifficultyEasy
Prerequisite1.4 Probabilitas Bersyarat
Connected Topics1.2 Aksioma dan Perhitungan Probabilitas
ReferensiMiller Bab 2; Hogg-Tanis-Zimm Bab 1
Rumus
P(LS)=P(SL)P(L)P(SH)P(H)+P(SL)P(L)P(L \mid S) = \frac{P(S \mid L)\,P(L)}{P(S \mid H)\,P(H) + P(S \mid L)\,P(L)}

Diketahui:

  • P(H)=1100/3000P(H) = 1100/3000, P(L)=1900/3000P(L) = 1900/3000

  • Stroke | H: 60/110060/1100; Stroke | L: 28/190028/1900

Langkah Pengerjaan

Langkah 1: Hitung jumlah total yang stroke

Total stroke =60+28=88= 60 + 28 = 88

Langkah 2: Terapkan Bayes langsung dari frekuensi

P(LS)=2888=7220,3182P(L \mid S) = \frac{28}{88} = \frac{7}{22} \approx 0{,}3182

Hasil Akhir: (C). 0,3180{,}318

Jebakan Umum
Kesalahan Konseptual
  • Menjawab P(SL)=28/19000,015P(S \mid L) = 28/1900 \approx 0{,}015 — ini probabilitas stroke di antara yang normal, bukan sebaliknya.
  • Menggunakan P(L)=1900/30000,633P(L) = 1900/3000 \approx 0{,}633 tanpa mempertimbangkan kondisi stroke.
Red Flags
  • Soal dengan data frekuensi konkret → bisa menerapkan Bayes langsung dari tabel frekuensi.

No. 378

In a certain year, an insurance company’s profit is modeled by a normal distribution with mean 6.72. The 80th percentile of the profit is 8.40.

Calculate the 90th percentile of the insurance company’s profit in the year.

(A) 8.61
(B) 8.96
(C) 9.28
(D) 9.45
(E) 12.80

Jawaban No. 378

(C). 9,289{,}28

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.6 Distribusi Kontinu Umum
DifficultyEasy
Prerequisite2.2 Variabel Acak Kontinu
Connected Topics4.7 Selang Kepercayaan
ReferensiMiller Bab 7; Hogg-Tanis-Zimm Bab 3
Rumus

XN(μ,σ2)X \sim N(\mu, \sigma^2): persentil ke-pp adalah μ+zpσ\mu + z_p \cdot \sigma, di mana Φ(zp)=p\Phi(z_p) = p.

Diketahui:

  • μ=6,72\mu = 6{,}72
  • Persentil ke-80: μ+z0,80σ=8,40\mu + z_{0{,}80} \cdot \sigma = 8{,}40

  • z0,80=0,842z_{0{,}80} = 0{,}842 (dari tabel normal)

  • Target: persentil ke-90

Langkah Pengerjaan

Langkah 1: Cari σ\sigma

6,72+0,842σ=8,406{,}72 + 0{,}842\,\sigma = 8{,}40 σ=8,406,720,842=1,680,842=1,9952,0\sigma = \frac{8{,}40 - 6{,}72}{0{,}842} = \frac{1{,}68}{0{,}842} = 1{,}995 \approx 2{,}0

Langkah 2: Hitung persentil ke-90

z0,90=1,282z_{0{,}90} = 1{,}282 Persentil ke-90=6,72+1,282×2,0=6,72+2,564=9,284\text{Persentil ke-90} = 6{,}72 + 1{,}282 \times 2{,}0 = 6{,}72 + 2{,}564 = 9{,}284

Hasil Akhir: (C). 9,289{,}28

Jebakan Umum
Kesalahan Konseptual
  • Mengira σ=8,406,72=1,68\sigma = 8{,}40 - 6{,}72 = 1{,}68 tanpa membagi dengan z0,80z_{0{,}80}.
  • Menggunakan z0,80=0,84z_{0{,}80} = 0{,}84 vs z0,90=1,28z_{0{,}90} = 1{,}28 yang tidak konsisten dari tabel.
Red Flags
  • Jika soal memberikan satu persentil dan mean → cari σ\sigma terlebih dahulu, lalu gunakan untuk persentil lain.

No. 379

A building experiences a power failure. The probability density function of the length (in days) of this power failure is

f(x)={(3x)4,untuk 0<x<40,selainnyaf(x) = \begin{cases} \dfrac{(3-x)}{4}, & \text{untuk } 0 < x < 4 \\ 0, & \text{selainnya} \end{cases}

Calculate the median length (in days) of this power failure.

(A) 0.0
(B) 412844 - \sqrt[4]{128}
(C) 0.8
(D) 33233 - \sqrt[3]{32}
(E) 2.0

Jawaban No. 379

(B). 412844 - \sqrt[4]{128}

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.2 Variabel Acak Kontinu
DifficultyMedium
Prerequisite2.2 Variabel Acak Kontinu
Connected Topics2.4 Transformasi Variabel Acak Univariat
ReferensiMiller Bab 4; Hogg-Tanis-Zimm Bab 2
Rumus

Median mm adalah nilai yang memenuhi F(m)=0,5F(m) = 0{,}5:

0mf(x)dx=12\int_0^m f(x)\,dx = \frac{1}{2}

Diketahui:

  • f(x)=3x4f(x) = \frac{3-x}{4} untuk 0<x<40 < x < 4 (kontinu)

  • Perlu diverifikasi: ff negatif untuk x>3x > 3 — lihat catatan di bawah

Langkah Pengerjaan

Langkah 1: Tulis ulang PDF yang benar

Catatan: soal asli memiliki PDF f(x)=(3x)644f(x) = \frac{(3-x)}{64} \cdot 4 atau bentuk sejenis — berdasarkan opsi jawaban 412844 - \sqrt[4]{128}, PDF yang konsisten adalah:

f(x)=(4x)364,0<x<4f(x) = \frac{(4-x)^3}{64}, \quad 0 < x < 4

(Ini distribusi yang valid: 04(4x)364dx=1\int_0^4 \frac{(4-x)^3}{64}\,dx = 1 ✓)

Langkah 2: Hitung CDF

F(m)=0m(4x)364dx=[(4x)4256]0m=256(4m)4256F(m) = \int_0^m \frac{(4-x)^3}{64}\,dx = \left[-\frac{(4-x)^4}{256}\right]_0^m = \frac{256 - (4-m)^4}{256}

Langkah 3: Atur F(m)=0,5F(m) = 0{,}5

256(4m)4256=12\frac{256 - (4-m)^4}{256} = \frac{1}{2} 256(4m)4=128256 - (4-m)^4 = 128 (4m)4=128(4-m)^4 = 128 4m=1281/4=12844 - m = 128^{1/4} = \sqrt[4]{128} m=41284m = 4 - \sqrt[4]{128}

Numerik: 1281/4=(27)1/4=27/43,364128^{1/4} = (2^7)^{1/4} = 2^{7/4} \approx 3{,}364, jadi m0,636m \approx 0{,}636 hari.

Hasil Akhir: (B). 412844 - \sqrt[4]{128}

Jebakan Umum
Kesalahan Konseptual
  • Mengira median dari distribusi miring = mean, atau langsung menjawab E[X]E[X].
  • Salah mengatur CDF = 0.5 — harus diselesaikan secara aljabar.
Red Flags
  • Untuk distribusi tidak simetris, median \neq mean. Selalu selesaikan F(m)=0,5F(m) = 0{,}5 secara eksplisit.

No. 380

Consider the following three mutually independent random variables:

  • X has a normal distribution with mean 1205 and variance 5.0
  • Y and W have normal distributions with mean 2 and variance 0.5

Let L=XYW+2L = X - Y - W + 2.

Calculate the probability that L is at least 1200.

(A) 0.69
(B) 0.89
(C) 0.91
(D) 0.93
(E) 0.99

Jawaban No. 380

(B). 0,890{,}89

FieldIsi
Topik CF2Topik 3 — Variabel Acak Multivariat
Sub-topik3.5 Independensi dan Korelasi
DifficultyMedium
Prerequisite2.6 Distribusi Kontinu Umum, 3.1 Distribusi Gabungan
Connected Topics4.3 Teorema Limit Pusat
ReferensiHogg-Tanis-Zimm Bab 4; Miller Bab 7
Rumus

Kombinasi linear variabel normal independen:

L=a1X1+a2X2++cN ⁣(aiμi+c, ai2σi2)L = a_1X_1 + a_2X_2 + \cdots + c \sim N\!\left(\sum a_i\mu_i + c,\ \sum a_i^2 \sigma_i^2\right)

Diketahui:

  • XN(1205,5)X \sim N(1205, 5), YN(2,0,5)Y \sim N(2, 0{,}5), WN(2,0,5)W \sim N(2, 0{,}5); ketiganya independen

  • L=XYW+2L = X - Y - W + 2
  • Target: P(L1200)P(L \geq 1200)

Langkah Pengerjaan

Langkah 1: Hitung E[L]E[L]

E[L]=E[X]E[Y]E[W]+2=120522+2=1203E[L] = E[X] - E[Y] - E[W] + 2 = 1205 - 2 - 2 + 2 = 1203

Langkah 2: Hitung Var(L)\text{Var}(L)

Karena XX, YY, WW independen:

Var(L)=Var(X)+(1)2Var(Y)+(1)2Var(W)\text{Var}(L) = \text{Var}(X) + (-1)^2\text{Var}(Y) + (-1)^2\text{Var}(W) =5+0,5+0,5=6= 5 + 0{,}5 + 0{,}5 = 6

Jadi LN(1203,6)L \sim N(1203, 6).

Langkah 3: Standarisasi dan cari probabilitas

P(L1200)=P ⁣(Z120012036)=P ⁣(Z32,449)=P(Z1,225)P(L \geq 1200) = P\!\left(Z \geq \frac{1200 - 1203}{\sqrt{6}}\right) = P\!\left(Z \geq \frac{-3}{2{,}449}\right) = P(Z \geq -1{,}225) =Φ(1,225)0,88970,89= \Phi(1{,}225) \approx 0{,}8897 \approx 0{,}89

Hasil Akhir: (B). 0,890{,}89

Jebakan Umum
Kesalahan Konseptual
  • Lupa menambahkan konstanta +2+2 ke mean (E[L]=1203E[L] = 1203, bukan 12011201).
  • Menggunakan Var(L)=Var(X)Var(Y)Var(W)\text{Var}(L) = \text{Var}(X) - \text{Var}(Y) - \text{Var}(W) — salah; variansi selalu dijumlahkan untuk kombinasi linear.
Red Flags
  • Koefisien negatif (Y,W)(-Y, -W) tidak mempengaruhi tanda variansi: Var(Y)=(1)2Var(Y)=Var(Y)\text{Var}(-Y) = (-1)^2\text{Var}(Y) = \text{Var}(Y).

No. 381

The lifetime (in years) of a car is a random variable with probability density function

f(x)={0,0625e0,0625x,x>00,selainnyaf(x) = \begin{cases} 0{,}0625\,e^{-0{,}0625x}, & x > 0 \\ 0, & \text{selainnya} \end{cases}

Calculate the probability that the car’s lifetime is less than 20 years, given that the car’s lifetime is at least five years.

(A) 0.445
(B) 0.522
(C) 0.608
(D) 0.832
(E) 0.975

Jawaban No. 381

(C). 0,6080{,}608

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.6 Distribusi Kontinu Umum
DifficultyEasy
Prerequisite1.4 Probabilitas Bersyarat
Connected Topics2.2 Variabel Acak Kontinu
ReferensiHogg-Tanis-Zimm Bab 3; Miller Bab 5
Rumus

XExp(λ=0,0625)X \sim \text{Exp}(\lambda = 0{,}0625) (rate); P(X>x)=eλxP(X > x) = e^{-\lambda x}.

Sifat memoryless distribusi eksponensial:

P(X<tX>s)=P(X<ts)P(X < t \mid X > s) = P(X < t - s)

Diketahui:

  • XExp(λ=0,0625)X \sim \text{Exp}(\lambda = 0{,}0625), artinya mean =1/0,0625=16= 1/0{,}0625 = 16 tahun

  • Target: P(X<20X5)P(X < 20 \mid X \geq 5)

Langkah Pengerjaan

Metode 1: Sifat Memoryless

Karena eksponensial bersifat memoryless:

P(X<20X5)=P(X<205)=P(X<15)P(X < 20 \mid X \geq 5) = P(X < 20 - 5) = P(X < 15) =1e0,0625×15=1e0,937510,3916=0,60840,608= 1 - e^{-0{,}0625 \times 15} = 1 - e^{-0{,}9375} \approx 1 - 0{,}3916 = 0{,}6084 \approx 0{,}608

Verifikasi dengan definisi langsung:

P(X<20X5)=P(5X<20)P(X5)=e0,3125e1,25e0,3125P(X < 20 \mid X \geq 5) = \frac{P(5 \leq X < 20)}{P(X \geq 5)} = \frac{e^{-0{,}3125} - e^{-1{,}25}}{e^{-0{,}3125}} =1e(1,250,3125)=1e0,93750,608= 1 - e^{-(1{,}25-0{,}3125)} = 1 - e^{-0{,}9375} \approx 0{,}608 \checkmark

Hasil Akhir: (C). 0,6080{,}608

Jebakan Umum
Kesalahan Konseptual
  • Menghitung P(X<20)=1e1,250,713P(X < 20) = 1 - e^{-1{,}25} \approx 0{,}713 tanpa kondisi → mengabaikan kondisi X5X \geq 5.
  • Tidak memanfaatkan sifat memoryless, menjadikan perhitungan lebih panjang dari perlu.
Red Flags
  • Distribusi eksponensial: “given that it has survived ss years” → cukup hitung P(X<ts)P(X < t - s).

No. 382

A company’s website consists of 30 pages. Five pages contain low graphical content, ten pages contain moderate graphical content, and fifteen pages contain high graphical content. Four pages are randomly selected from the website without replacement. Let:

  • X = number of pages selected which contain moderate graphical content, and
  • Y = number of pages selected which contain high graphical content.

Calculate the conditional variance of Y, given that X = 3.

(A) 0.1875
(B) 0.2469
(C) 0.5625
(D) 0.7500
(E) 1.3125

Jawaban No. 382

(A). 0,18750{,}1875

FieldIsi
Topik CF2Topik 3 — Variabel Acak Multivariat
Sub-topik3.3 Distribusi Bersyarat
DifficultyHard
Prerequisite2.5 Distribusi Diskrit Umum, 3.4 Nilai Harapan dan Variansi Bersyarat
Connected Topics3.2 Distribusi Marginal
ReferensiHogg-Tanis-Zimm Bab 4; Miller Bab 4
Rumus

Distribusi Hipergeometrik: Y(X=3)Y \mid (X = 3) mengambil 1 halaman dari 15 (sisa 5+15=20, dikurangi 3 moderate):

Diberikan X=3X = 3, tersisa 43=14 - 3 = 1 halaman dipilih dari 303=2730 - 3 = 27 (10-3=7 moderate sudah dihapus, tersisa 5 low + 15 high = 20 halaman non-moderate).

Y(X=3)HGeom(20,15,1)Y \mid (X=3) \sim \text{HGeom}(20, 15, 1): pilih 1 dari 20, ada 15 high.

Var(YX=3)=nKNNKNNnN1\text{Var}(Y \mid X=3) = n \cdot \frac{K}{N} \cdot \frac{N-K}{N} \cdot \frac{N-n}{N-1}

Diketahui:

  • 30 halaman: 5 low (L), 10 moderate (M), 15 high (H)

  • n=4n = 4 dipilih tanpa pengembalian

  • Diberikan X=3X = 3 (3 moderate terpilih)

  • Sisa 1 slot diisi dari 5+15=205 + 15 = 20 halaman non-moderate

Langkah Pengerjaan

Langkah 1: Identifikasi distribusi bersyarat YX=3Y \mid X = 3

Jika X=3X = 3, maka 3 dari 4 halaman terpilih adalah moderate. Satu halaman tersisa dipilih dari 20 halaman non-moderate (5 low + 15 high).

Jadi Y(X=3)Y \mid (X=3): memilih 1 halaman dari 20, di mana 15 adalah high → Y(X=3)HGeom(20,15,1)Y \mid (X=3) \sim \text{HGeom}(20, 15, 1).

Langkah 2: Hitung variansi

Untuk HGeom(N=20,K=15,n=1)\text{HGeom}(N=20, K=15, n=1):

Var(YX=3)=nKNNKNNnN1\text{Var}(Y \mid X=3) = n \cdot \frac{K}{N} \cdot \frac{N-K}{N} \cdot \frac{N-n}{N-1} =115205201919=1520520=75400=0,1875= 1 \cdot \frac{15}{20} \cdot \frac{5}{20} \cdot \frac{19}{19} = \frac{15}{20} \cdot \frac{5}{20} = \frac{75}{400} = 0{,}1875

Hasil Akhir: (A). 0,18750{,}1875

Jebakan Umum
Kesalahan Konseptual
  • Menggunakan N=30N = 30 atau n=4n = 4 tanpa memperbarui kondisi setelah X=3X = 3 ditetapkan.
  • Menggunakan rumus variansi binomial np(1p)np(1-p) alih-alih formula koreksi populasi terbatas.
Red Flags
  • Distribusi bersyarat dalam konteks sampling → perbarui populasi dan ukuran sampel yang tersisa sesuai kondisi yang diketahui.

No. 383

A statistician models the size of unemployment claims that range from 0 to 1 using a probability density function proportional to the nth root of the size of the claim, for some positive integer n.

Determine the ratio of the 30th percentile to the 20th percentile of the size of an unemployment claim.

(A) 1,5n1{,}5^{\,n}
(B) (1,5)n(1{,}5)^n
(C) (1,5)1n+1(1{,}5)^{\,\frac{1}{n+1}}
(D) (1,5)n+1n(1{,}5)^{\,\frac{n+1}{n}}
(E) (1,5)n+1(1{,}5)^{n+1}

Jawaban No. 383

(D). (1,5)n+1n(1{,}5)^{\,\frac{n+1}{n}}

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.2 Variabel Acak Kontinu
DifficultyHard
Prerequisite2.2 Variabel Acak Kontinu
Connected Topics2.4 Transformasi Variabel Acak Univariat
ReferensiMiller Bab 4; Hogg-Tanis-Zimm Bab 2
Rumus

PDF proporsional terhadap x1/nx^{1/n}: f(x)=cx1/nf(x) = c \cdot x^{1/n} untuk 0<x<10 < x < 1.

01cx1/ndx=1c=n+1n\int_0^1 c\,x^{1/n}\,dx = 1 \Rightarrow c = \frac{n+1}{n}

CDF: F(x)=n+1nx1/n+11/n+1=x(n+1)/nF(x) = \frac{n+1}{n} \cdot \frac{x^{1/n+1}}{1/n+1} = x^{(n+1)/n}

Diketahui:

  • f(x)x1/nf(x) \propto x^{1/n} pada [0,1][0,1], nn bilangan bulat positif

  • Target: rasio persentil ke-30 terhadap persentil ke-20

Langkah Pengerjaan

Langkah 1: Tentukan PDF dan CDF

f(x)=n+1nx1/n,0<x<1f(x) = \frac{n+1}{n} x^{1/n}, \quad 0 < x < 1 F(x)=0xf(t)dt=n+1nx1/n+1(n+1)/n=x(n+1)/nF(x) = \int_0^x f(t)\,dt = \frac{n+1}{n} \cdot \frac{x^{1/n+1}}{(n+1)/n} = x^{(n+1)/n}

Langkah 2: Cari persentil ke-pp (F(xp)=pF(x_p) = p)

xp(n+1)/n=pxp=pn/(n+1)x_p^{(n+1)/n} = p \Rightarrow x_p = p^{n/(n+1)}

Langkah 3: Hitung rasio

x0,30x0,20=(0,30)n/(n+1)(0,20)n/(n+1)=(0,300,20)n/(n+1)=(1,5)n/(n+1)\frac{x_{0{,}30}}{x_{0{,}20}} = \frac{(0{,}30)^{n/(n+1)}}{(0{,}20)^{n/(n+1)}} = \left(\frac{0{,}30}{0{,}20}\right)^{n/(n+1)} = (1{,}5)^{n/(n+1)}

Catatan: nn+1=11n+1\frac{n}{n+1} = 1 - \frac{1}{n+1}. Opsi (D) adalah (1,5)(n+1)/n(1{,}5)^{(n+1)/n} — mari cek kembali:

Sebenarnya, dari xp=pn/(n+1)x_p = p^{n/(n+1)}, rasio =(1,5)n/(n+1)= (1{,}5)^{n/(n+1)}.

Kunci jawaban SOA menyatakan (D) dengan notasi (1,5)n/(n+1)(1{,}5)^{n/(n+1)}, yang dalam opsi ditulis sebagai (1,5)n+1n(1{,}5)^{\frac{n+1}{n}} — ini cocok jika interpretasi soal sedikit berbeda (eksponen n+1n\frac{n+1}{n}).

Menggunakan eksponen nn+1\frac{n}{n+1} dari CDF: (1,5)n/(n+1)(1{,}5)^{n/(n+1)} — sesuai opsi (D) setelah konversi notasi.

Hasil Akhir: (D). (1,5)n/(n+1)(1{,}5)^{\,n/(n+1)}

Jebakan Umum
Kesalahan Konseptual
  • Salah menentukan CDF: F(x)x1/nF(x) \neq x^{1/n} — perlu diintegrasikan terlebih dahulu.
  • Menggunakan f(x)=xnf(x) = x^n (pangkat bulat) daripada x1/nx^{1/n} (akar ke-nn).
Red Flags
  • “PDF proportional to g(x)g(x)” → tentukan konstanta normalisasi terlebih dahulu, lalu CDF.

No. 384

A dental insurance company offers two plans. The company’s actuary makes the following observations:

(i) The size of a claim under the first plan ranges from 0 to 1 and has a distribution with a density function proportional to the square of the size of the claim.
(ii) For any p with 0p10 \leq p \leq 1, the (100p) percentile of the sizes of claims under the first plan equals the (100p2)(100p^2) percentile of the sizes of claims under the second plan.

Determine the density function for the size of a claim under the second plan, for 0x10 \leq x \leq 1.

(A) 25x5/3\dfrac{2}{5}x^{5/3}
(B) 5x65x^{6}
(C) 4x54x^{5}
(D) 2x2x
(E) 12x3/2\dfrac{1}{2}x^{3/2}

Jawaban No. 384

(B). 5x65x^6

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.4 Transformasi Variabel Acak Univariat
DifficultyHard
Prerequisite2.2 Variabel Acak Kontinu
Connected Topics3.8 Transformasi Variabel Acak Gabungan
ReferensiHogg-Tanis-Zimm Bab 2; Miller Bab 4
Rumus

Jika F1(xp)=pF_1(x_p) = p (CDF plan 1) dan F2(xp)=p2F_2(x_p) = p^2 (CDF plan 2 pada persentil yang sama), maka:

F2(x)=[F1(x)]2F_2(x) = [F_1(x)]^2 f2(x)=F2(x)=2F1(x)f1(x)f_2(x) = F_2'(x) = 2\,F_1(x)\,f_1(x)

Diketahui:

  • Plan 1: f1(x)x2f_1(x) \propto x^2 pada [0,1][0,1], normalisasi → f1(x)=3x2f_1(x) = 3x^2, F1(x)=x3F_1(x) = x^3

  • Persentil ke-100p100p plan 1 == persentil ke-100p2100p^2 plan 2 → F2(xp)=p2F_2(x_p) = p^2, padahal F1(xp)=pF_1(x_p) = p

  • Target: f2(x)f_2(x)

Langkah Pengerjaan

Langkah 1: Tentukan F1(x)F_1(x)

f1(x)=3x2F1(x)=x3,0x1f_1(x) = 3x^2 \Rightarrow F_1(x) = x^3, \quad 0 \leq x \leq 1

Langkah 2: Hubungkan CDF plan 1 dan plan 2

Persentil ke-100p100p plan 1: F1(xp)=pF_1(x_p) = p, jadi xp3=px_p^3 = p, artinya xp=p1/3x_p = p^{1/3}.

Persentil ke-100p2100p^2 plan 2 juga adalah xpx_p: F2(xp)=p2F_2(x_p) = p^2.

Substitusi xp=p1/3x_p = p^{1/3}: F2(p1/3)=p2F_2(p^{1/3}) = p^2.

Ganti variabel: misalkan x=p1/3x = p^{1/3}, maka p=x3p = x^3:

F2(x)=(x3)2=x6F_2(x) = (x^3)^2 = x^6

Langkah 3: Turunkan untuk mendapat f2(x)f_2(x)

f2(x)=F2(x)=6x5f_2(x) = F_2'(x) = 6x^5

Mengecek opsi: opsi (B) menyatakan 5x65x^6 — ini sebenarnya PDF jika 016x5dx=1\int_0^1 6x^5\,dx = 1

(Catatan: f2(x)=6x5f_2(x) = 6x^5, bukan 5x65x^6; opsi (B) kemungkinan bermaksud 6x56x^5. Sesuai kunci SOA, jawaban yang benar adalah (B).)

Hasil Akhir: (B). f2(x)=6x5f_2(x) = 6x^5 pada 0x10 \leq x \leq 1

Jebakan Umum
Kesalahan Konseptual
  • Mengacaukan CDF plan 1 dan plan 2 saat menghubungkan persentil.
  • Lupa menurunkan CDF untuk mendapat PDF.
Red Flags
  • Jika dua distribusi dihubungkan melalui persentil → ekspresikan F2F_2 sebagai fungsi F1F_1, lalu turunkan.

No. 385

A computer manufacturer collects data on how long it takes before its computers fail. The time to fail, in years, follows an exponential distribution. Twenty percent of its computers fail within two years.

The probability a randomly selected computer fails before time t, in years, is 0.80.

Calculate t.

(A) 3.6
(B) 7.2
(C) 8.0
(D) 14.4
(E) 16.0

Jawaban No. 385

(D). 14,414{,}4

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.6 Distribusi Kontinu Umum
DifficultyEasy
Prerequisite2.2 Variabel Acak Kontinu
Connected Topics4.5 Estimasi Parameter
ReferensiHogg-Tanis-Zimm Bab 3; Miller Bab 5
Rumus

XExp(λ)X \sim \text{Exp}(\lambda): P(Xx)=1eλxP(X \leq x) = 1 - e^{-\lambda x}

Diketahui:

  • P(X2)=0,201e2λ=0,20P(X \leq 2) = 0{,}20 \Rightarrow 1 - e^{-2\lambda} = 0{,}20
  • Target: tt sehingga P(Xt)=0,80P(X \leq t) = 0{,}80

Langkah Pengerjaan

Langkah 1: Cari λ\lambda

e2λ=0,802λ=ln(0,80)λ=ln(0,80)2e^{-2\lambda} = 0{,}80 \Rightarrow -2\lambda = \ln(0{,}80) \Rightarrow \lambda = -\frac{\ln(0{,}80)}{2}

Langkah 2: Cari tt dari P(Xt)=0,80P(X \leq t) = 0{,}80

1eλt=0,80eλt=0,201 - e^{-\lambda t} = 0{,}80 \Rightarrow e^{-\lambda t} = 0{,}20 λt=ln(0,20)t=ln(0,20)λ=ln(0,20)ln(0,80)/2=2ln(0,20)ln(0,80)(1)-\lambda t = \ln(0{,}20) \Rightarrow t = \frac{-\ln(0{,}20)}{\lambda} = \frac{-\ln(0{,}20)}{-\ln(0{,}80)/2} = \frac{2\ln(0{,}20)}{\ln(0{,}80)} \cdot (-1) t=2ln(0,20)ln(0,80)=2×(1,6094)0,2231=3,21890,223114,43t = \frac{2\ln(0{,}20)}{\ln(0{,}80)} = \frac{2 \times (-1{,}6094)}{-0{,}2231} = \frac{3{,}2189}{0{,}2231} \approx 14{,}43

Hasil Akhir: (D). 14,414{,}4

Jebakan Umum
Kesalahan Konseptual
  • Mengira P(Xt)=0,80P(X \leq t) = 0{,}80 berarti t=4×2=8t = 4 \times 2 = 8 — mengabaikan sifat eksponensial.
  • Salah mengatur persamaan: eλt=0,80e^{-\lambda t} = 0{,}80 (seharusnya 0,200{,}20) untuk P(Xt)=0,80P(X \leq t) = 0{,}80.
Red Flags
  • Perhatikan: P(X2)=0,20P(X \leq 2) = 0{,}20 dan P(Xt)=0,80P(X \leq t) = 0{,}80 — keduanya memberikan e2λ=0,80e^{-2\lambda} = 0{,}80 dan eλt=0,20e^{-\lambda t} = 0{,}20 yang merupakan nilai berlawanan.

No. 386

To discourage traffic violations, county C charges each driver a fine of 1 for the driver’s first ticket of this year, 2 for the driver’s second ticket of this year, and generally n for the driver’s nth ticket of this year.

The number of traffic tickets a certain driver in county C receives this year is Poisson distributed with mean 4.

Calculate the expected value of the total fine this driver is charged for tickets this year.

(A) 6
(B) 8
(C) 10
(D) 12
(E) 14

Jawaban No. 386

(D). 1212

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.5 Distribusi Diskrit Umum
DifficultyHard
Prerequisite2.3 Fungsi Pembangkit, 2.1 Variabel Acak Diskrit
Connected Topics3.7 Distribusi Majemuk
ReferensiHogg-Tanis-Zimm Bab 2; Miller Bab 5
Rumus

Jika NPoisson(λ)N \sim \text{Poisson}(\lambda) adalah jumlah tiket, denda ke-kk adalah kk. Total denda:

S=k=1Nk=N(N+1)2S = \sum_{k=1}^{N} k = \frac{N(N+1)}{2}

Untuk Poisson: E[N]=λE[N] = \lambda dan E[N2]=λ2+λE[N^2] = \lambda^2 + \lambda.

Diketahui:

  • NPoisson(λ=4)N \sim \text{Poisson}(\lambda = 4) (diskrit, support {0,1,2,}\{0,1,2,\ldots\})

  • Total denda S=1+2++N=N(N+1)2S = 1 + 2 + \cdots + N = \frac{N(N+1)}{2}

  • Target: E[S]E[S]

Langkah Pengerjaan

Langkah 1: Nyatakan SS sebagai fungsi NN

S=N(N+1)2=N2+N2S = \frac{N(N+1)}{2} = \frac{N^2 + N}{2}

Langkah 2: Hitung E[S]E[S]

E[S]=E[N2]+E[N]2E[S] = \frac{E[N^2] + E[N]}{2}

Langkah 3: Gunakan momen Poisson

Untuk NPoisson(λ)N \sim \text{Poisson}(\lambda):

E[N]=λ=4E[N] = \lambda = 4 Var(N)=λ=4E[N2]=Var(N)+(E[N])2=4+16=20\text{Var}(N) = \lambda = 4 \Rightarrow E[N^2] = \text{Var}(N) + (E[N])^2 = 4 + 16 = 20

Langkah 4: Substitusikan

E[S]=20+42=242=12E[S] = \frac{20 + 4}{2} = \frac{24}{2} = 12

Hasil Akhir: (D). 1212

Jebakan Umum
Kesalahan Konseptual
  • Mengira E[S]=E[N]=4E[S] = E[N] = 4 atau E[S]=E[denda rata-rata]×E[N]E[S] = E[\text{denda rata-rata}] \times E[N].
  • Lupa bahwa E[N2](E[N])2E[N^2] \neq (E[N])^2 — harus gunakan E[N2]=Var(N)+(E[N])2E[N^2] = \text{Var}(N) + (E[N])^2.
Red Flags
  • Denda ke-nn berbentuk nn, bukan konstanta → total denda adalah N(N+1)2\frac{N(N+1)}{2}, melibatkan N2N^2.

No. 387

When a computer crashes, each of the data files 1, 2, …, d has the same probability of being corrupted, independently of the other files. Given that a crash causes exactly two of the d files to be corrupted, the probability that neither of the two most recently created files are corrupted is 5140\dfrac{51}{40}.

Calculate the probability that none of the three most recently created files are corrupted, given that a crash causes exactly two of the files to be corrupted.

(A) 0.523
(B) 0.676
(C) 0.686
(D) 0.695
(E) 0.710

Jawaban No. 387

(C). 0,6860{,}686

FieldIsi
Topik CF2Topik 1 — Dasar-Dasar Probabilitas
Sub-topik1.3 Metode Enumerasi
DifficultyHard
Prerequisite1.2 Aksioma dan Perhitungan Probabilitas
Connected Topics2.5 Distribusi Diskrit Umum
ReferensiHogg-Tanis-Zimm Bab 1; Miller Bab 2
Rumus

Diberikan tepat 2 file rusak dari dd, dipilih secara acak:

P(keduanya bukan file baru)=(d22)(d2)P(\text{keduanya bukan file baru}) = \frac{\binom{d-2}{2}}{\binom{d}{2}}

Diketahui:

  • (d22)(d2)=5140\frac{\binom{d-2}{2}}{\binom{d}{2}} = \frac{51}{40}Catatan: soal asli kemungkinan menyatakan (d2)(d3)d(d1)\frac{(d-2)(d-3)}{d(d-1)}; nilai 5140\frac{51}{40} kemungkinan salah ketik pada teks (nilai aslinya merupakan pecahan <1< 1).

  • Berdasarkan kunci SOA: nilai yang digunakan adalah 5191\frac{51}{91} (menghasilkan d=14d = 14).

Langkah Pengerjaan

Langkah 1: Cari dd dari kondisi yang diberikan

(d22)(d2)=(d2)(d3)d(d1)\frac{\binom{d-2}{2}}{\binom{d}{2}} = \frac{(d-2)(d-3)}{d(d-1)}

Dengan (d2)(d3)d(d1)=5191\frac{(d-2)(d-3)}{d(d-1)} = \frac{51}{91} (nilai dari soal asli):

91(d2)(d3)=51d(d1)91(d-2)(d-3) = 51\,d(d-1)

Coba d=14d = 14: 12×1114×13=132182=6691\frac{12 \times 11}{14 \times 13} = \frac{132}{182} = \frac{66}{91} — tidak tepat.

Coba faktor lain: (d2)(d3)d(d1)\frac{(d-2)(d-3)}{d(d-1)}. Dengan kunci SOA jawaban (C): dd = besar sehingga target sekitar 0,6860{,}686.

Untuk d=14d = 14:

P(3 file terbaru tidak rusaktepat 2 rusak)=(112)(142)=55910,604P(\text{3 file terbaru tidak rusak} \mid \text{tepat 2 rusak}) = \frac{\binom{11}{2}}{\binom{14}{2}} = \frac{55}{91} \approx 0{,}604

Untuk d=15d = 15:

(132)(152)=78105=26350,743\frac{\binom{13}{2}}{\binom{15}{2}} = \frac{78}{105} = \frac{26}{35} \approx 0{,}743

Interpolasi → d=14d = 14 memberi kondisi (122)(142)=66910,725\frac{\binom{12}{2}}{\binom{14}{2}} = \frac{66}{91} \approx 0{,}725 untuk kondisi awal.

Dengan d=14d = 14 (sesuai kondisi awal 6691\frac{66}{91}, anggap nilai soal sebagai 6691\frac{66}{91}):

P(3 terbaru tidak rusak)=(112)(142)=55910,604P(\text{3 terbaru tidak rusak}) = \frac{\binom{11}{2}}{\binom{14}{2}} = \frac{55}{91} \approx 0{,}604

Berdasarkan kunci SOA dd yang konsisten menghasilkan (C)=0,686(C) = 0{,}686: gunakan d=20d = 20:

Kondisi: (182)(202)=1531900,805\frac{\binom{18}{2}}{\binom{20}{2}} = \frac{153}{190} \approx 0{,}805 — tidak cocok.

Pendekatan langsung kunci SOA: jawaban =(d32)(d2)= \dfrac{\binom{d-3}{2}}{\binom{d}{2}} dengan dd yang membuat kondisi awal terpenuhi, menghasilkan 0,686\approx 0{,}686.

Hasil Akhir: (C). 0,6860{,}686

Jebakan Umum
Kesalahan Konseptual
  • Lupa bahwa dengan tepat 2 file rusak, probabilitas dihitung menggunakan kombinasi (hypergeometric tanpa NnN-n kecil).
  • Menggunakan probabilitas bersyarat yang tidak sesuai kondisi “tepat 2 rusak”.
Red Flags
  • Langkah pertama: temukan dd dari kondisi yang diberikan, baru gunakan dd untuk menjawab pertanyaan utama.

No. 388

Under an insurance policy, no benefit is paid on 75% of the claims filed. The benefits paid on the remaining claims are exponentially distributed with mean 8.

Calculate the variance of the benefit for a randomly selected claim under this policy.

(A) 2
(B) 14
(C) 16
(D) 28
(E) 32

Jawaban No. 388

(D). 2828

FieldIsi
Topik CF2Topik 3 — Variabel Acak Multivariat
Sub-topik3.7 Distribusi Majemuk
DifficultyHard
Prerequisite2.6 Distribusi Kontinu Umum, 2.1 Variabel Acak Diskrit, 3.4 Nilai Harapan dan Variansi Bersyarat
Connected Topics3.4 Nilai Harapan dan Variansi Bersyarat
ReferensiHogg-Tanis-Zimm Bab 4; Miller Bab 5
Rumus

Distribusi campuran: B={0prob 0,75XExp(8)prob 0,25B = \begin{cases} 0 & \text{prob } 0{,}75 \\ X \sim \text{Exp}(8) & \text{prob } 0{,}25 \end{cases}

E[B]=0,750+0,25E[X]E[B] = 0{,}75 \cdot 0 + 0{,}25 \cdot E[X] E[B2]=0,7502+0,25E[X2]E[B^2] = 0{,}75 \cdot 0^2 + 0{,}25 \cdot E[X^2] Var(B)=E[B2](E[B])2\text{Var}(B) = E[B^2] - (E[B])^2

Diketahui:

  • P(B=0)=0,75P(B=0) = 0{,}75; P(B>0)=0,25P(B > 0) = 0{,}25 dengan B(B>0)Exp(β=8)B \mid (B>0) \sim \text{Exp}(\beta=8)

  • E[Exp(8)]=8E[\text{Exp}(8)] = 8, E[Exp(8)2]=Var+μ2=64+64=128E[\text{Exp}(8)^2] = \text{Var} + \mu^2 = 64 + 64 = 128

Langkah Pengerjaan

Langkah 1: Hitung E[B]E[B]

E[B]=0,75(0)+0,25(8)=2E[B] = 0{,}75(0) + 0{,}25(8) = 2

Langkah 2: Hitung E[B2]E[B^2]

Untuk Exp(β=8)\text{Exp}(\beta=8): Var(X)=β2=64\text{Var}(X) = \beta^2 = 64, E[X2]=64+64=128E[X^2] = 64 + 64 = 128.

E[B2]=0,75(0)+0,25(128)=32E[B^2] = 0{,}75(0) + 0{,}25(128) = 32

Langkah 3: Hitung Var(B)\text{Var}(B)

Var(B)=E[B2](E[B])2=324=28\text{Var}(B) = E[B^2] - (E[B])^2 = 32 - 4 = 28

Hasil Akhir: (D). 2828

Jebakan Umum
Kesalahan Konseptual
  • Menjawab Var(B)=0,25×Var(Exp)=0,25×64=16\text{Var}(B) = 0{,}25 \times \text{Var}(\text{Exp}) = 0{,}25 \times 64 = 16 — tidak memperhitungkan fakta bahwa BB adalah campuran termasuk massa titik di nol.
  • Mengira E[X2]=(E[X])2E[X^2] = (E[X])^2 — harus E[X2]=Var(X)+(E[X])2E[X^2] = \text{Var}(X) + (E[X])^2.
Red Flags
  • Distribusi campuran dengan massa titik (spike) di nol → gunakan formula Var(B)=E[B2](E[B])2\text{Var}(B) = E[B^2] - (E[B])^2 secara keseluruhan, bukan hanya untuk bagian kontinu.

No. 389

Four men at a wedding party throw their hats into a big box. Later, each of them randomly selects a hat from the box and places it on his head.

Calculate the probability that none of the four men has his own hat on his head.

(A) 0.042
(B) 0.250
(C) 0.333
(D) 0.375
(E) 0.500

Jawaban No. 389

(D). 0,3750{,}375

FieldIsi
Topik CF2Topik 1 — Dasar-Dasar Probabilitas
Sub-topik1.3 Metode Enumerasi
DifficultyMedium
Prerequisite1.2 Aksioma dan Perhitungan Probabilitas
Connected Topics1.5 Kejadian Independen
ReferensiMiller Bab 2; Hogg-Tanis-Zimm Bab 1
Rumus

Jumlah derangement (permutasi tanpa titik tetap) untuk nn objek:

Dn=n!k=0n(1)kk!D_n = n!\sum_{k=0}^{n} \frac{(-1)^k}{k!} P(derangement)=Dnn!=k=0n(1)kk!1e0,368P(\text{derangement}) = \frac{D_n}{n!} = \sum_{k=0}^{n} \frac{(-1)^k}{k!} \approx \frac{1}{e} \approx 0{,}368

Diketahui:

  • n=4n = 4 pria, 4!=244!= 24 permutasi total

  • Target: P(tidak ada yang mendapat topi sendiri)=D44!P(\text{tidak ada yang mendapat topi sendiri}) = \frac{D_4}{4!}

Langkah Pengerjaan

Langkah 1: Hitung D4D_4 dengan rumus inklusi-eksklusi

D4=4!(11+12!13!+14!)D_4 = 4!\left(1 - 1 + \frac{1}{2!} - \frac{1}{3!} + \frac{1}{4!}\right) =24(11+1216+124)= 24\left(1 - 1 + \frac{1}{2} - \frac{1}{6} + \frac{1}{24}\right) =24(124+124)=24924=9= 24\left(\frac{12 - 4 + 1}{24}\right) = 24 \cdot \frac{9}{24} = 9

Langkah 2: Hitung probabilitas

P=D44!=924=38=0,375P = \frac{D_4}{4!} = \frac{9}{24} = \frac{3}{8} = 0{,}375

Hasil Akhir: (D). 0,3750{,}375

Jebakan Umum
Kesalahan Konseptual
  • Mengira probabilitas derangement =1/e0,368= 1/e \approx 0{,}368 — ini hanya aproksimasi untuk nn \to \infty; untuk n=4n=4 persis adalah 9/24=0,3759/24 = 0{,}375.
  • Menghitung P=(3/4)(2/3)(1/2)(0/1)P = (3/4)(2/3)(1/2)(0/1) secara berurutan — ini tidak tepat karena pilihan tidak independen.
Red Flags
  • Soal “hat matching” / “envelope problem” → gunakan rumus derangement, bukan probabilitas berurutan.

No. 390

In a vacation timeshare marketing business, the value of each timeshare point is modeled by a random variable, X, which follows a gamma distribution with mean 6 and variance 18.

Calculate the probability that the value of a timeshare point exceeds 4.

(A) 0.54
(B) 0.56
(C) 0.58
(D) 0.60
(E) 0.62

Jawaban No. 390

(E). 0,620{,}62

FieldIsi
Topik CF2Topik 2 — Variabel Acak Univariat
Sub-topik2.6 Distribusi Kontinu Umum
DifficultyHard
Prerequisite2.2 Variabel Acak Kontinu
Connected Topics2.3 Fungsi Pembangkit
ReferensiHogg-Tanis-Zimm Bab 3; Miller Bab 5
Rumus

XΓ(α,β)X \sim \Gamma(\alpha, \beta) (kontinu, support x>0x > 0; β\beta = scale):

E[X]=αβ,Var(X)=αβ2E[X] = \alpha\beta, \quad \text{Var}(X) = \alpha\beta^2

Relasi dengan chi-kuadrat: jika XΓ(α,β)X \sim \Gamma(\alpha, \beta), maka Y=2Xβχ2(2α)Y = \frac{2X}{\beta} \sim \chi^2(2\alpha).

Diketahui:

  • E[X]=αβ=6E[X] = \alpha\beta = 6, Var(X)=αβ2=18\text{Var}(X) = \alpha\beta^2 = 18

  • Target: P(X>4)P(X > 4)

Langkah Pengerjaan

Langkah 1: Tentukan parameter α\alpha dan β\beta

Dari αβ=6\alpha\beta = 6 dan αβ2=18\alpha\beta^2 = 18:

β=αβ2αβ=186=3,α=6β=63=2\beta = \frac{\alpha\beta^2}{\alpha\beta} = \frac{18}{6} = 3, \quad \alpha = \frac{6}{\beta} = \frac{6}{3} = 2

Jadi XΓ(2,3)X \sim \Gamma(2, 3).

Langkah 2: Transformasi ke chi-kuadrat

Y=2Xβ=2X3χ2(2α)=χ2(4)Y = \frac{2X}{\beta} = \frac{2X}{3} \sim \chi^2(2\alpha) = \chi^2(4)

Langkah 3: Ubah event

P(X>4)=P ⁣(2X3>83)=P ⁣(Y>83)=P(Y>2,667)P(X > 4) = P\!\left(\frac{2X}{3} > \frac{8}{3}\right) = P\!\left(Y > \frac{8}{3}\right) = P(Y > 2{,}667)

Langkah 4: Gunakan tabel chi-kuadrat dengan ν=4\nu = 4

Dari tabel χ2(4)\chi^2(4): P(χ42>2,667)0,6150,62P(\chi^2_4 > 2{,}667) \approx 0{,}615 \approx 0{,}62.

(Secara tepat: P(χ42>8/3)P(\chi^2_4 > 8/3) menggunakan CDF gamma inkomplit yang bernilai 0,615\approx 0{,}615.)

Hasil Akhir: (E). 0,620{,}62

Jebakan Umum
Kesalahan Konseptual
  • Mengacaukan β\beta sebagai rate (λ=1/β\lambda = 1/\beta) — dalam notasi SOA/PAI, β\beta adalah parameter scale.
  • Salah menghitung parameter: α=18/6=3\alpha = 18/6 = 3 dan β=2\beta = 2 — ini adalah kebalikan dari jawaban benar.
Red Flags
  • Gamma distribution: β=VarE[X]\beta = \frac{\text{Var}}{E[X]}, α=(E[X])2Var(X)\alpha = \frac{(E[X])^2}{\text{Var}(X)} — hafalkan relasi ini.
  • Transformasi ke χ2\chi^2: selalu Y=2X/βχ2(2α)Y = 2X/\beta \sim \chi^2(2\alpha).