INSTITUTE OF ACTUARIES OF INDIA
EXAMINATIONS
28th November 2023
Subject CS1B – Actuarial Statistics (Paper B)
Time allowed: 1 Hour 45 Minutes (10.15 – 12.00 Hours)
Total Marks: 100
INSTRUCTIONS TO THE CANDIDATES
1. Mark allocations are shown in brackets.
2. Attempt all questions beginning your answer to each question on a new page.
3. Attempt all sub-parts of the question in one document only, unless otherwise
instructed to do so.
4. All the detailed guidelines are available on exam screen.
5. Do save your work in solution template on a regular basis.
6. If Any, Data set file(s) accompanying the question paper is available for download on
the exam screen.
7. You need to import the same into R studio as soon as you begin the exam.
8. Ensure to copy and paste R codes and output at regular intervals onto the solution
template.
9. Please check if you have received complete Question Paper and no page is missing.
If so, kindly get new set of Question Paper from the Invigilator.
AT THE END OF THE EXAMINATION
Please return this question paper to the supervisor separately. You are not allowed to carry the
question paper in any form with you. You are requested to save and submit the work before leaving
the examination premises.
IAI CS1B-1123
Q. 1)
i) Hotel room prices in Mumbai are normally distributed with a mean of Rs. 5400 and
standard deviation of Rs. 900, whereas prices in Kolkata are normally distributed with
a mean of Rs. 3600 and standard deviation of Rs. 1500. Compute the probability that a
hotel room price in Mumbai is atleast twice the price in Kolkata. Note that the hotel
prices in Mumbai and Kolkata are independent of each other.
Store mean and sd values as x.mean, x.sd, y.mean,y.sd,etc.
Hint: p**** is a useful function to ascertain CDF. (5)
ii)
a) Show that the population mean and standard deviation of “differences in hotel prices
between Mumbai and Kolkata” (i.e., Mumbai hotel prices – Kolkata hotel Prices)
are 1800 and 1749.86. (3)
b) Generate a sample of size 50 for the ” differences in hotel prices between Mumbai
and Kolkata”. (2)
c) Draw qqplot and qqline for the sample generated and comment on the results. (4)
Store population mean and standard deviation of differences as dif.mean and dif.sd
Make sure to set the seed value to 1234 i.e use set.seed(1234) before generating the sample.
Store the sample as dif.sample
Don’t paste the sample
iii) Test using 5% significance level whether the mean of “difference in hotel prices” is less
than 1375 based on the sample generated in part (ii).
a) Perform z-test and comment. (5)
b) Perform t-test (assuming population is not known) and comment. (3)
Please ensure to include test conclusion.
iv)
a) Generate another sample but with size=1000 and store as dif.sample2. (1)
b) Draw qqplot and qqline for the sample generated in (iv.a) and comment on the
results. (3)
[26]
Q. 2) You are a Research Analyst and working on a short term assignment. Your professor
provided dance data (dance.csv) containing scores of 20 contestants of a famous dance show
and asked you to perform the following tasks.
i) Using read.csv load the data and use head command to view the first few rows of the (2)
data.
No need to paste the output of head command.
dance.csv contains following fields:
- Judges: It indicates the score provided by judges
- Audience: It indicates the sum of scores provided by audience
Page 2 of 6
IAI CS1B-1123
- Final: It indicates the final score of the contestant
ii) Plot a scatterplot for each pair of data. (4)
Make sure to paste the scatterplot in your answer scripts.
iii) Using scatterplot of part (ii), comment on the relationship between the pairs of data. (3)
iv) Fit a multiple linear regression model with
- Final score as response variable and
- Judges and Audience score as explanatory variables.
a) Store the model as m1. (2)
b) Show summary of the output and also write the equation of the fitted model. (4)
c) Compute confidence interval of all parameters. (3)
d) Comment on the significance of the each explanatory variable either referring to
above confidence interval or other statistics. (3)
You published the report showing the above results and reference data on your research
website. After few months, a journalist accused the dance show of ignoring audience scores
and cited your report.
The sponsors of dance show stated that incomplete data is used for analysis as audience
score depends upon the number of audiences provided scores. They shared supplement data
providing the number of audiences for each participant.
v) Using the following code to load the number of audiences for 20 contestants
audience.count<-c(110,100,90,120,100,100,100,100,110,110,100,
100,110,90,100,110,120,120,100,100)
Also, verify total count is equal to 2090. (2)
vi) Compute the new audience score by dividing the sum of score provided by audience (2)
(“Audience”) with audience count (“audience.count”).
Store as Audience2 and make sure to attach to the dance data.
Please don’t paste Audience2 values.
vii) Perform correlation test to check whether any correlation exits between final score and
new audience score. (3)
viii) Fit a new multiple linear regression model with
- Final score as response variable and
- Judges and New Audience score as explanatory variables.
Store this model as m2 and show the summary of the output. (3)
Page 3 of 6
IAI CS1B-1123
ix) Using a suitable statistic from part (iv.b) and (viii). outputs, compare models m1 and (3)
m2 and suggest which model is better.
Please write the figures of the statistic while answering.
[34]
Q. 3) An analyst fitted various models on the data containing claims information of 20 policies
and shared the output to conduct analysis.
Output 1:
Call:
glm(formula = Claim ~ 1, family = poisson(lin = "log"), data = q3)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.8439 -0.8975 -0.1791 0.3925 2.5561
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 0.5306 0.1715 3.094 0.00197 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for poisson family taken to be 1)
Null deviance: 30.147 on 19 degrees of freedom
Residual deviance: 30.147 on 19 degrees of freedom
AIC: 71.114
Number of Fisher Scoring iterations: 5
i) Write the equation of the fitted model using Output 1 as above and specify which
distribution is used to model the response variable. (2)
ii) Using output 1, show that the sample mean of the response variable is 1.7 (when
rounded to one decimal place). (2)
Output 2:
Call:
glm(formula = Claim ~ Gender + Health - 1, family = poisson(lin = "log"),
data = q3)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.22474 -0.81754 -0.07119 0.27453 1.44149
Coefficients:
Estimate Std. Error z value Pr(>|z|)
GenderF 1.1394 0.3490 3.265 0.001096 **
GenderM 1.1394 0.2216 5.143 2.71e-07 ***
HealthNonDiabetic -1.4271 0.3939 -3.623 0.000291 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Page 4 of 6
IAI CS1B-1123
(Dispersion parameter for poisson family taken to be 1)
Null deviance: 38.229 on 20 degrees of freedom
Residual deviance: 14.436 on 17 degrees of freedom
AIC: 59.403
Number of Fisher Scoring iterations: 5
iii) Write the equation of the fitted model using Output 2. (4)
Make sure to define the explanatory (categorical) variables.
iv) Output 2 doesn’t have an intercept as a coefficient unlike output1. Please provide the
reason for the same by comparing glm R formulas. (2)
v) Compare output 1 and output 2, and suggest which model is better using suitable
statistics. (2)
vi) Compute log likelihood of the model given in output 2. (3)
vii) Claim follows Poisson distribution. Test for the parameter to be equal to 1.5 at 1%
level of significance. (5)
Hint: Use outputs and various sub-parts to determine x for poisson test.
[20]
Q. 4) The table below shows the total claim number (cancellations) per year, Xij, for 4 travel
companies over last 4 years.
Years,j
1 2 3 4
Companies, i
Make 455 458 587 531
Travel
Ease 251 322 292 340
Go 309 246 217 120
Clear 400 426 470 547
i) Using Empirical Bayes Credibility Theory (EBCT) Model 1, compute the following
a) Copy the below code to load the data: (1)
q4<-matrix(c(455,251,309,400,
458,322,246,426,
587,292,217,470,
531,340,120,547),
ncol=4,nrow=4)
b) E[m(θ)] (2)
c) E[s2(θ)] (2)
d) Var[m(θ)] (3)
e) Z (2)
Page 5 of 6
IAI CS1B-1123
ii) Using part (i), calculate the expected claim number for Go and Clear. (3)
iii) What additional information is required to use EBCT Model 2. (2)
iv) Travel company “Ease” launched a membership program last year providing full
refund on cancellations. Number of cancellations believed to follow binomial
distribution with parameters n=3 and 0.20.
Number of cancellations in last year on 150 memberships are as follows:
Cancellations 0 1 2 3
Members 61 71 15 3
Carry out goodness of fit test for the binomial model specified for number of cancellations
on each membership. (5)
[20]
****************************
Page 6 of 6
