Course: APFFIN407_2023 – Mathématiques financières Deadline:

Professor: Francesco SANNA 01 September 2024, 23.59 CEST

Student: Annaëlle BILOUNGA

DEVOIR MAISON
----- Instructions -----
The devoir maison is an individual assignment and it consists of 4 parts. The student is kindly asked to
upload a pdf file with the solution written at the computer. Files written by hand or photos will not be
graded.

Late submission penalty.

Each 24 hours of delay represents a 5% decrease in the final grade, i.e. the grade will be reduced by 1/20 pt
for each day of delay. Should the devoir maison be uploaded after September 4th, 2024, 23.59 CEST time,
the devoir maison is considered as if it was not submitted at all.

Example: if a student uploads the file on September 2 nd, 00.10 CEST and the devoir maison was perfect,
s/he will get 19/20 instead of 20/20. If another student will upload the file on September 3 rd, 00.01 and the
devoir maison is perfect, s/he will get 18/20. If a student uploads the devoir maison on September 5th, 2024,
00.00 CEST or later, the final grade will be 0/20.

Use of external sources and AI usage policy.

Given this is a personal assignment, students are not allowed to use any information from internet, any AI
tool to solve problems, nor to collaborate with other students. By uploading the file, students are taking full
responsibility for the content and are declaring that they are aware that AI usage is not permitted. If it is
verified that external sources or AI tools have been used for one or more exercises, the exercise(s) will be
counted as not being submitted.

----- Content -----

Q1. [max 5/20 pt]

In two years from now you are supposed to make 3 annual payments of €10000 at the end of the next
three years. How much do you need to annually save at the end of each year to have the amount requested
by the beginning of the third year?

If you have a net annual salary of €15000, would you be able to pay the 3 amounts? Comment on it.

Assume that the rate of return is 5%.

[Evaluation. 1/20 is assigned for the timeline. 3/20 pt are assigned for the right assumptions and calculations. 1/20 pt is attributed
for the commentary]

Q2. [max 5/20 pt]

You have been provided with the following data:

14% 11% 6% 13% 12% 10% 8% 5% 130% 10.5%

 Q2.1 (2/20 pt). The student is asked to share all calculations for the arithmetic mean (0.5/20 pt),
standard deviation (0.5/20 pt), skewness (0.5/20 pt) and kurtosis (0.5/20 pt).

Q2.2 (1/20 pt). After copying the provided data on an Excel file, using Excel formulas, the student is
kindly asked to calculate median (0.25/20 pt), range (0.25/20 pt), 25 th percentile (0.25/20 pt) and 75th
percentile (0.25/20 pt). You are asked to upload the excel file. If only the final values are provided in the
Excel file, this part will be graded as 0/20 pt.

Should the student not be able to calculate it with the software, s/he is asked to calculate it by hand,
although this won’t be valued.

Q2.3 (2/20 pt). The student is kindly asked to comment on each quantity calculated (1/20 pt), as
well as to provide an interpretation of data and to suggest what s/he would do to manage data properly, in
order to provide a meaningful interpretation (1/20 pt).

Q3. [max 5/20 pt]

A student of Audencia arrives at the faculty 70% of times by metro. In 80% cases the student is sharp on
time at the classes of financial mathematics. On average the student is on time at school 60% of times.

Today the student has arrived on time at school. What is the probability that the student has taken the
metro?
[Evaluation. Points are assigned on the basis of the correct assumptions (2/20 pt) and of the proper use of mathematical tools (1/20
pt for theory and 2/20 pt for the calculations).]

Q4. [max 5/20 pt]

You have been provided with the following data, where X is the production and Y the prices. You need to
build a simple model. In particular your supervisor has asked you to build a simple regression line, using the
least squares method. In addition, each student is asked to comment about the quality of the regression
line.
[Evaluation. For the solution, the student is asked to share the calculations of all quantities needed to build the model (3/20 pt).
1/20 is assigned for the graph. 1/20 is assigned for the commentary on the data quality].

2
 -----Answers-----

Q1. Payment due

- In two years from now, the payment will be made at the end of year 3,4
and 5:
NOW (0) 1 2 3 4 5

Beginning of that year

If the annual rate of return is 5%, the amount of money needed to be saved at the
end of each year to have the amount requested by the beginning of the third year
are:
Payment at the end of year 3:
PV = 10 000/(1,05)^0 = 10000
At the end of the year 4:
PV = 10 000/(1,05)^1 = 9523,81
Payment At the end of year 5:
PV = 10 000/(1,05)^2 = 9070,29

Meaning that 28 594,10€ to be saved by the beginning of year 3 to make all the 3
annual payments considering that the rate of return is 5%. So, we need to
determine the amount of money the person has to save in the two years before the
beginning of the third year.

3
With a rate of return of 5%, assuming the person is going to save the same amount
each year, the annual savings each two years is:
If A = annual saving we have,
A(1,05)^0 + A(1,05)^1 = 28584,10
A + A (1,05) + A(1,1025) = 28584,10
A (1 + 1,05) = 28584,10
A*2,05 = 28584,10
A = 28594,10/2,05
A = 13 948,34 €

With an annual salary of 15 000€, assuming the person live independently in an EU

(European Union) country, based on the currency of the exercise. If the person saves
the require annual savings of 13 948,34 €, only 1 051,66 € will remain for other
expenses such as living expenses (housing, foods, and so on) which is most likely not
enough to live with for a year in any EU country. Therefore, only based on their
salary the person wouldn’t have enough money to have the required amount to
make the payment.

Q2.1. Statistic data calculation

14% 11% 6% 13% 12% 10 8% 5% 130% 10.5%
%

Results are rounded to the nearest tenth.

Assuming the given data is a sample of a population,
- Arithmetic mean
If M = the arithmetic mean of the data,

4
 N

M= ∑ xi
i=1
N −1
With N = number of counts, x = data

M = (14 + 11 + 6 + 13 + 12 + 10 + 8 + 5 + 130 + 10,5)/10

M = 21,95%
- Standard deviation
Let’s calculate the variance first.
If V = variance,
N

V = ∑ ¿¿ ¿ ¿
i=1

V = [ (14 – 21,95)^2 + (11-21,95)^2 + (6 -21,95)^2 + (13-21,95)^2 + (12-21,95)^2 +

(10-21,95)^2 + (8-21,95)^2 + (5-21,95)^2 + (130-21,95)^2 + (10,5-21,95)^2] / (10 – 1)
V = 1449,692
If SD = standard deviation,
SD = √ V
SD = √ 1449,692
SD = 38,07

- Skewness
If SK = skewness,
N

SK = ∑ ¿¿ ¿ ¿
i=1

SK = [ (14 – 21,95)^3 + (11-21,95)^3 + (6 -21,95)^3 + (13-21,95)^3 + (12-21,95)^3 +

(10-21,95)^3 + (8-21,95)^3 + (5-21,95)^3 + (130-21,95)^3 + (10,5-21,95)^3] / ((10 –
1) 38,07482*^3)

SK = 1243095 / ((10 – 1)*38,07482^3)

SK = 1243095 / (9 - 55196,74)

5
SK = 3,13
- Kurtosis
If K = kurtosis,
N

K= ∑ ¿¿ ¿ ¿
i=1

K = [ (14 – 21,95)^4 + (11-21,95)^4 + (6 -21,95)^4 + (13-21,95)^4 + (12-21,95)^4 +

(10-21,95)^4 + (8-21,95)^4 + (5-21,95)^4 + (130-21,95)^4 + (10,5-21,95)^4] / ((10 –
1)*

38,07482^4)
K = 136558316,4 / (9*2101606)
K = 9,84
Q2.2 Excel
On Excel
- Median
Input: =MEDIAN(A2:A11)
Result: 10,75
- Range
Input: =SMALL(A2:A11;1)
Result: 5 in cell F3
Input: =LARGE(A2:A11;1)
Result: 130 in cell G3
Input: =G3 – F3
6
Result: 125
- 25th percentile
Input: =PERCENTILE(A2:A11;0,25)
Result: 8,5
- 75th percentile
Input: =PERCENTILE(A2:A11;0,75)
Result: 12,75

Q2.3. Interpretation
A mean of 21,95 when most data are under 20, indicates that the average is highly
influence by the outlier which is 130. A high standard deviation of 38.07 indicates
that the data points are widely dispersed around the mean, again primarily due to
the outlier.
The skewness of this dataset is 3,13 indicating that it is highly skewed to the right (it
is a positive skew), still due to the presence of the large outlier. Kurtosis of 9,84
means a distribution with heavy tails and sharp peak (higher than the normal
distribution)
10.75 for the median indicates that half of the data points are below this value and
half are above it. The median is often a better measure of central tendency than the
average in the presence of an outlier. A range of 125 indicates significant spread in
the data, largely driven by the presence of the outlier value of 130.
The 25th percentile is 8,5 indicating that 25% of the data is below this point. The 75th
percentile is 12,75, indicating that 75% of the data values are under this point, which
marks even more the influence of the large outlier value (130) on the data
distribution.
Overall, the descriptive statistic of this data distribution clearly shows that one large
value can highly influence the results and may mislead in decision making. I believe
to have a better observation of this dataset, the extraordinary value (130) should be
removed or maybe the sample should be bigger, indeed they are only 10 values in
this dataset, it is quite limited especially when there is an outlier value in the set.

7
Q3. Probability
We need to know what is the probability that the student used the metro
considering that s/he arrived on time today. Here, we need to calculate a
conditional probability using the Bayes theorem:
With the following events:
M = arrives by metro
(T/M) = arrives on time using the metro – conditional
M/T = used the metro considering that s/he arrived on time
T = arrives on time (average)
And their probabilities:
P(M) = 0,7
P(T/M) = 0,8
P(T) = 0,6
P(M/T) = ?

P(M/T) = P(T/M) * P(M) / P(T)

P(M/T) = (0,8*0,7)/0,6
P(M/T) = 0,93
The probability that the student has taken the metro given that he arrived on time
today is 93%.
Q4. Linear Regression line
Using the least square method, we have the following formula:
Y = a + bX
Where,
Y = Production
X = Price
a = intercept (price expected on average when production = 0)
b = slope (for how much the price change when one the production change by one)
8
Now, we have an equation with two unknown values (a and b), that we must find to
draw to linear regression line.
- Calculating the slope of the line (b)

∑ ( x−xm )∗( y− ym)

b=
∑ (x−xm¿)2 ¿
where x = production, y = price, xm = average of production, ym = average of price
So, we have,
Production Price (y) (x – xm) (y – ym) (x-xm)*(y- (x-xm)^2
(x) ym)
30 25 1,4* -10,4** -14,56*** 1,96****
28 30 -0,6 -5,4 3,24 0,36
32 27 3,4 -8,4 -28,56 11,56
25 40 -3,6 4,6 -16,56 12,96
25 42 -3,6 6,6 -23,76 12,96
25 40 -3,6 4,6 -16,56 12,96
22 50 -6,6 14,6 -96,36 43,56
24 45 -4,6 9,6 -44,16 21,16
35 30 6,4 -5,4 -34,56 40,96
40 25 11,4 -10,4 -118,56 129,96
Total 286 354 -1,4 1,4 -3,9 2,9

xm = 286/10 = 28,6
ym = 354/10 = 35,4
*30 – 28,6 = 1,4
**25-35,4 = -10,4
***1,4*-10,4 = -14,56
****1,4^2 = 1,96
b = -3,9/2,9
b = - 1,35 (rounded to the nearest hundredth)

9
 - Finding the intercept (a)
Now that we have b, we have an equation with one unkown values that we can
easily resolve :
Y = a + bX
Using the averages and calculation of b,
35,4 = a -1,35*28,6
35,4 = a -38,61
a = 35,4 + 38,61
a = 74,01

So, the formula we have is :

Y = 74,01 – 1,35X

Production Price (y) Y = 74,01 -1,35X

(x)
30 25 33,51*
28 30 36,21
32 27 30,81
25 40 40,26
25 42 40,26
25 40 40,26
22 50 44,31
24 45 41,61
35 30 26,76
40 25 20,01

*74,01 -1,35*30 = 33,51

Using Excel, the graph of this regression line is :

10
An intercept of 74,01 means that the production is egal to zero, the price of the
product can be expected to be 74,01 on average.
The slope is negative. X and Y have an inverse relationship. When the production
increases, the price of the product decreases. When the production increases by 1,
the price decreases by 1,35.
Overall, the dots are following the trend, except some outliers that most likely are
influenced by other variable.
The negative relationship between the price and the production suggests that this
regression line is a representation of the economies of scale, where increasing
production lower the cost of the additional units produced and therefore lower the
price of the product.

Note: Compared to Excel, the intercept is slightly different because of rounding when calculating the slope.
On Excel, it is 74,11.

11
12