Institute and Faculty of Actuaries: Subject CT6 - Statistical Methods Core Technical
Institute and Faculty of Actuaries: Subject CT6 - Statistical Methods Core Technical
Institute and Faculty of Actuaries: Subject CT6 - Statistical Methods Core Technical
EXAMINERS REPORT
April 2015 examinations
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
Page 2
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
(i)
(ii)
A saddle point exists if an entry is both the largest in its column and the
smallest in its row.
This can only occur in row 2 (the smallest values in rows 1 and 3 are not the
largest in their columns).
X cannot give a saddle since this would require X 8 (to be the smallest in
row 2) but then X would not be the largest in its column.
Equally Y cannot give a saddle point as this would require that Y 8 in which
case Y would not be the largest in its column.
So there are no values of X and Y which give a saddle point.
This straightforward question was relatively poorly answered, with many candidates
seemingly put off by the unfamiliar nature of the question.
2207
= 1.0480
2106
DF1,2 =
(2106 2985)
= 1.2253
(1969 2186)
DF0,1 =
Page 3
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
= 3752.16 1
= 1452.01
Total claims = 1452.01 + 1773
= 3225.01
This standard chain ladder question was very well answered by the majority of candidates.
(i)
(a)
(b)
(ii)
= 270 and
= 3402
2
1
( 1) ( 2)
= 3402
2
2 ( 1)
so
3402
=
= 1.585733882
2
2702
so =
2 1.585733882
1.585733882 1
= 5.4145
and = 270 4.4145 = 1191.920375
Page 4
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
i.e.
= 0.05
M
0.05
1
0.05 ( M )
1 0.05
M =
1
0.05
1
5.4145
1191.920375 1 0.05
=
1
0.05 5.4145
= 880.8
This question was the best answered on the paper with most candidates scoring well. Some
candidates were unable to manipulate the Pareto distribution.
(i)
Page 5
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
(ii)
= max
h( x )
f ( x)
2 xe x
x 0
e x
= max 2 xe x x
x 0
log g(x)
= log2 + logx + x x2
d log g ( x)
dx
= 1 1 2x
x
i.e.
1 + x 2x2 = 0
2x2 x 1 = 0
(2x + 1)(x 1) = 0
x = 1 or x =
2
h( x )
= xe x x
2 f ( x)
Page 6
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
(iii)
1
1
=
so on average 2 simulations are needed to
c
2
return 1 value, so in this case we need 20 simulations. Each simulation
requires 2 pseudo-random numbers so on average we will need 40 pseudorandom numbers.
We accept a proportion
Most candidates were able to produce good quality answers to part (i), although the majority
struggled with part (ii). Candidates with the confidence to apply the acceptance-rejection
method scored well.
P1 = 17 + 23 + 21 + 29 + 35 = 125
P2 = 42 + 51 + 60 + 55 + 37 = 245
P3 = 43 + 31 + 62 + 98 + 107 = 341
P = 125 + 245 + 341 = 711
X =
Z1 =
so
25, 200
= 840
30
840 829.18
= 0.519594
850 829.18
125
E ( s 2 ())
125
Var[m()]
= 0.51969
so
E ( s 2 ())
125 0.51969 125
=
= 115.57217
Var[m()]
0.51969
so
Z2 =
245
= 0.6794756
245 115.528
Z3 =
341
= 0.74686
341 115.528
Page 7
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
Candidates with good knowledge of EBCT Model 2 scored well here, however a
disappointing number of candidates were apparently unfamiliar with this method.
(i)
l=
e 41 11
1
4
yij !
j 1
so
L
11
4
1
and setting
i.e.
11
=0
1
1 =
11
= 2.75
4
similarly
3 =
Page 8
L
= 0 we have
1
2 =
16
=4
4
20
=5
4
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
(ii)
Assuming = 1 = 2 = 3 we have =
11 16 20
= 3.916667
12
(i)
(ii)
Page 9
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
(iii)
12 = (12 1212 )2 12 (1 12 ) 2
13 112 2
and
so
1 =
2
1 (1 12
)
2
(1 12 )(1 12
)
11 = 13 =
12 =
1
1 12
112
2
(1 12 )(1 12
)
12 (1 12 )
2
(1 12 )(1 12
)
12
2
1 12
Page 10
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
(i)
First note that N has a type 2 negative binomial distribution with parameters
p = 0.8 and k = 1. Hence
E(N) =
0.2
= 0.25
0.8
Var(N) =
0.2
0.82
= 0.3125
1000
=
= 250
4
1
Var(X) = 2502
5
= 104,166.666 = (322.75)2
3
(a)
= P N (0,1)
213.478
2
2
= 62.5
e 2 (e 1) = 213.4782
2
(A)
(B)
Page 11
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
(B) (A)2 e 1 =
2
213.4782
62.52
= 11.66665
= P N (0,1)
1.5934
This question was well answered by the majority of candidates. Full credit was given to
alternative correct answers in part (iii).
(i)
0 p g ( x ) p
L(g(x)p) =
1 otherwise
The expected loss is
E(L) = 1
g
g
f ( p x)dp
= 1 2f(gx)
Page 12
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
34 = (208 g ) p
p =
34
208 g
Under quadratic loss the Bayes estimate is given by mean of the posterior
distribution which is given by
p =
35
35
=
35 175 g 210 g
Page 13
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
i.e.
i.e.
i.e.
10
(i)
= dt
1 x
e dx
x
= dt
e M
= dt e
M
=
e
dt
Page 14
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
(ii)
M X i (t ) = E (etX i )
x
et ( x M ) 1 e dx
M
= e
1 e
0t
= 1 e
1 Mt x t
e e
dx
= 1 e
= 1 e
= 1 e
= 1 e
= 1 e
=1 e
(iii)
(a)
1
x t
e
e Mt
1 t
1
1
M t
e Mt
e
1 t
1
1
1 t
1 t
1
1 t
t
1 t
Now M X i (t ) = 1 e
So
M X' i (t )
=e
and so E(Xi) =
t
1 t
2
1 t (1 t )
M X' i (0) =
Page 15
Subject CT6 (Statistical Methods Core Technical) April 2015 Examiners Report
(b)
(iv)
1 + (1 + ) E(Xi)R = M X i ( R)
R = 1 e
i.e.
1 + (1 + ) e
i.e.
(1 + ) R =
i.e.
(1 + ) R(1 R) = R
i.e.
(1 + ) (1 R) = 1
i.e.
1 + R (R) = 1
i.e.
R(1 + ) =
i.e.
R=
R
1 R
R
1 R
(1 )
Page 16