COLLEGE OF COMPUTING, INFORMATICS AND

MATHEMATICS
BACHELOR OF SCIENCE (HONS.) MATHEMATICS
(CS249)
INTRODUCTION TO MATHEMATICAL MODELLING
(MAT530)
TITLE:
MATHEMATICAL MODELLING OF UNEMPLOYMENT IN
ASIAN COUNTRIES IN 2021
GROUP ASSIGNMENT
GROUP 4
PREPARED FOR:
DR. NOOREHAN BINTI AWANG
SUBMISSION DATE:
24 DECEMBER 2023
NO. NAME STUDENT ID GROUP
1 MUHAMMAD AFIF BIN 2021486374 N4CS2495A2
MOHAMAD SOBRI
2 MUHAMMAD ALIF HAIKAL 2021839902 N4CS2495A2
BIN MOHD AZNI
3 MUHAMMAD ASHRAF IMAN 2021826386 N4CS2495A2
BIN MD KAHARULNIZAM
Table of Contents

1.0 INTRODUCTION ..................................................................................................... 1

2.0 PROBLEM STATEMENT ........................................................................................ 2
3.0 OBJECTIVES ............................................................................................................ 2
4.0 LITERATURE REVIEW .......................................................................................... 3
4.1 MATHEMATICAL MODELLING ................................................................... 3
4.2 STATIC MODEL ............................................................................................... 3
4.3 APLLICATION OF MODEL ............................................................................ 4
5.0 METHODOLOGY .................................................................................................... 5
5.1 STATIC MODEL ............................................................................................... 6
5.1.1 LINEAR EQUATION ................................................................................ 6
5.1.2 QUADRATIC EQUATION ..................................................................... 10
5.1.3 SINGLE EXPONENTIAL EQUATION ................................................. 14
5.2 RESULT AND COMPARISON ...................................................................... 18
6.0 CONCLUSION ........................................................................................................ 18
7.0 REFERENCES ........................................................................................................ 18
1.0 INTRODUCTION

Unemployment occurs when an individual actively seeks employment but cannot secure a job,
serving as an indicator of the economic well-being. Meanwhile, unemployment in Asian
countries is a multifaceted economic phenomenon that reflects the dynamics of the labor
market across the continent. It encompasses the condition where individuals who are willing
and able to work find themselves without employment opportunities. As a crucial indicator of
economic health, unemployment in Asia is influenced by a myriad of factors, including but not
limited to economic policies, global market trends, technological advancements, and regional
variations in labor markets.

Mathematical modeling involves representing a real-world system through mathematical

equations, enabling a deeper understanding and the identification of novel aspects related to
the problem. Subsequently, these equations play a crucial role in explaining the system,
examining the effects of different components, and making predictions about the future. The
key components of mathematical models include assumptions, encompassing variables,
parameters, and functional forms. These models serve as valuable tools for comprehending,
analysing, and foreseeing the dynamics of complex systems.

In the Static model, the Least Squares method was employed. The Least Squares model
constitutes a form of mathematical regression analysis utilised to identify the optimal line for
a dataset, potentially illustrating the connection between data points visually. Each data point
signifies the link between a known independent variable and an unknown dependent variable.
Within the Least Squares method, three equations are utilised, including a linear equation,
polynomial equation, and exponential equation. The linear equation denotes the standard
relationship between the x-coordinates and y-coordinates of any point on the line.
Subsequently, a polynomial equation is characterized by an equation with multiple terms
composed of integers and variables. Lastly, an exponential equation takes the form of a single
exponent equation where the exponent is a variable.

The Discrete Dynamical System is characterised as a nonlinear system wherein a country

undergoes discrete time steps within a state space, following a predetermined rule. Among the
various models within Discrete Dynamical equations, the Discrete Malthusian Growth model
stands out as particularly popular. Named after the 18th-century British philosopher Thomas
Malthus, this model is frequently employed in ecological modeling. Nonlinear systems, such
as the Malthusian Growth model, find common application in understanding population
dynamics, especially in ecological contexts. This model, rooted in the principles articulated by
Thomas Malthus in 1766, posits that the population of the succeeding generation is directly
proportional to the population of the current generation. This conceptual framework is widely
utilised in predicting and comprehending population trends in diverse species or groups of
species, leveraging information about the preceding generation (MacRae, 2018).

The most used metric for measuring unemployment is the unemployment rate, which is
determined by dividing the number of unemployed individuals by the total working population
or the labour force (UN ESCAP, 2003). The dataset collected is about the unemployment rate
for 31 years of each country around the world i.e. from the year 1991-2021 but for this project
specifically used on unemployment in Asian countries in 2021. It will be using the method of
Static model within the Least Squares method on three equations are utilised which are linear
equation, quadratic equation, and exponential equation. Thus, this project will use
mathematical models to understand and explore the application of equations or functional
relationships.

2.0 PROBLEM STATEMENT

The overarching problematics addressed in this comprehensive study revolve around the
intricate and multifaceted nature of unemployment models within the Asian context during the
pivotal year of 2021. Firstly, the investigation aims to discern and elucidate the complex
mathematical equations underpinning each distinct model employed to articulate and
comprehend the intricate dynamics of unemployment prevalent in diverse Asian countries
during the aforementioned temporal epoch. Subsequently, the research endeavors to
proactively tackle the challenge of model identification by delving into the expansive realm of
data analytics, seeking to discern and articulate a systematic approach that facilitates the
recognition and substantiation of the optimal model, thereby contributing to a nuanced
understanding of the multifarious facets of unemployment patterns in the Asian landscape.
Lastly, the study grapples with the quintessential question of model appropriateness, aspiring
to unravel the most pertinent and applicable model for interpreting and forecasting the
unemployment data specific to Asian countries throughout the year 2021. In essence, this
research embarks on a multifaceted journey to unravel the mathematical intricacies, optimize
model selection processes, and discern the most fitting analytical framework for
comprehending the nuanced tapestry of unemployment across the Asian economic landscape
in the pivotal year of 2021.

3.0 OBJECTIVES

The objectives of this study are:

1. To identify the equations for each model of unemployment in Asian countries

in 2021.

2. To investigate the values of the Root Mean Square Deviation (RMSD) for
each model.

3. To determine the most suitable model for the data unemployment Asian
countries in 2021.
4.0 LITERATURE REVIEWS

4.1 MATHEMATICAL MODELLING

For a model to function, the information must be transformed into mathematical formulas,
including the data, concepts, relationships, conditions, and presumptions. Consequently, a
mathematical model of the initial situation appears. This kind of model consists of a particular
mathematical object that represents the original element/situation or a real-world model that
represents a relationship. More precisely, a mathematical model may consist of multiple real-
world issues, sets of mathematical entities, and relations where the object and relation are
connected to each other. Even though mathematics is a process from a real model into
mathematics, they use modelling or model construction to refer to the entire process that goes
from the original real problem scenario to the mathematical model. It has demonstrated efficacy
in differentiating between different types of models. If economic entities, like taxes or benefits,
are considered mathematical, it's primarily to set up particular regulations that involve value
assessments. It all boils down to the model of norms. Mathematics is mainly used to describe
and explain the conditions of physical phenomena like radioactive decay and planetary motion.
Here, descriptive models are everything (Russ, 2011).

4.2 STATIC MODEL

An investigation into the estimated impact of the COVID-19 lockdown in Spain on the size
and survival of skin tumours was conducted by Tejera-Vaquerizo et al. (2020). The goal of this
study was to model the effect of tumour growth delays on patient survival for melanoma and
squamous cell carcinoma (SCC). To assess the growth of tumours between the time of patient-
reported onset and surgical excision at various points in time, the researchers built an
exponential growth model for both SCC and melanoma. Large, thick SCCs and melanomas are
consequently becoming more common after the lockdown postponed the necessary surgery.

In their research, Sadiq and Qureshi (2010) aim to use linear and quadratic models to examine
the trends of two important climatic factors in Pakistan: rainfall and temperature. The
temperature trends for the maximum, minimum, and mean are determined using these two
models. Consequently, they discovered that the linear model's negative slope and excessively
high MAPE, MAD, and RMSD values make it difficult to calculate the rainfall in a number of
cities. Since the quadratic model produces marginally better results, they conclude that it is the
best model. Pakistan's major cities experience extremely non-linear rainfall patterns, so it is
important to use more advanced methods to identify regional trends, like Autoregressive
Integrated Moving Average (ARIMA) modelling and neural networks. Moreover, windowing
and space-time separation plots can be used to deal with the nonlinearities in the data.

As per the findings of a linear static analysis, pier height increases resulted in an increase in
positive bending moment, tensile stress, and downward vertical deflection values, as clarified
by Naser et al. (2021). Bridge structures will eventually collapse due to a decrease in their
structural performance, which is measured by their elasticity, carrying capacity of their
structural parts, and stiffness, but their compressive stresses and negative bending moment
were kept to a minimum. Therefore, high-strength concrete, more steel reinforcement, more
prestressed tendons, and larger girder and pier cross sections must all be used in the safe design
of bridge structures that make use of tall piers. The results of the modal analysis indicate that,
for three distinct bridge model types, the unloaded dynamic frequency decreased as the pier
heights increased, indicating a decrease in the stiffness of the bridge structure as the pier height
increased. The results of the time history analysis and response spectra show that the dynamic
displacement and loaded dynamic frequency (vibration state) increased with pier height. This
suggests that there will be significant vibration on the bridge of the structure at high pier
heights. The results of the study indicate that when bridge structures are subjected to traffic
stresses, the height of the piers has a significant impact on both the static and dynamic structural
performance of the structures.

4.3 APPLICATION OF MODEL

Analysis of continuum robots requires the use of static models. Analysis and static modelling
of continuum surgical robots were investigated by Yuan et al. (2016). This study discusses the
static modelling and analysis of single backbone continuum robots using the Newton-Euler
method. External loading and the effects of gravity are both taken into account. This static
model can be used to determine the robot's profile by providing controlling cable forces and
external loads. Additionally, the finite element method is employed as a tool to validate the
model that is provided. Investigations are conducted into the relationship between the external
loads, driven cable forces, and robot end position. External loading significantly affects the
continuum robots' static profile. Under external loads, the continuous curvature assumption is
no longer valid. For example, they found an S-type bending under specific loading conditions
using simulations based on the suggested model. The suggested model is crucial for ongoing
robot analysis, as it can improve the accuracy of static positioning under varied loading
conditions.

The research on gendered variations in travel behaviour in developed nations was expanded by
Chamseddine and Boubkr (2021) through an analysis of the variability of travel behaviour
within and between gender and income groups in the context of Casablanca city. According to
data from the 2018 Casablanca Travel Survey, women travel more for household maintenance
and fewer for work-related purposes overall than men, but these differences vary depending on
income level. Women typically take more trips than men when their income rises; the opposite
is true for middle-class and low-class individuals. According to the results, walking is the most
popular mode of transportation for both men and women in the lowest income groups, but in
the highest income groups, private cars have the largest modal share because both sexes avoid
nonmotorised modes of transportation as household income rises.

"Error propagation theory is used to estimate static formation temperatures in geothermal and
petroleum boreholes" was the topic of a study by Verma et al (2006). Using error propagation
theory, the researchers in this study estimated the uncertainty in static formation temperature
estimates in geothermal and petroleum wells. Thus, using the common least squares linear
regression model used in this work, they tested two versions of the weighted least squares linear
regression model for the true relationship between bottom hole temperature and the associated
time function. Equations based on error propagation theory were developed to assess the degree
of uncertainty in the time function of each analytical method. A weighted least squares linear
regression model offers a temperature estimate with less uncertainty than a standard least
squares linear regression model. Consequently, the weighted model would be statistically
correct and more suitable for the given application.
5.0 METHODOLOGY
Based on our time series data, the dataset of unemployment rate for 31 years of each country
around the world i.e. from the year 1991-2021 were taken from the websites, i.e. Kaggle
(https://www.kaggle.com/datasets/pantanjali/unemployment-dataset). This project specifically
used on unemployment in Asian countries in 2021 (Pant, 2022).

Asian country 2021

Afghanistan 13.28
Armenia 20.9
Azerbaijan 6.58
Bahrain 1.87
Bangladesh 5.23
Bhutan 4.33
Brunei Darussalam 7.65
Cambodia 0.61
China 4.82
Egypt, Arab Rep. 9.33
Georgia 10.66
Hong Kong SAR, China 5.32
India 5.98
Indonesia 4.41
Iran, Islamic Rep. 11.46
Iraq 14.19
Israel 5.05
Japan 2.8
Jordan 19.25
Kazakhstan 4.9
Kuwait 3.71
Kyrgyz Republic 9.1
Lao PDR 1.26
Lebanon 14.49
Macao SAR, China 3.01
Malaysia 4.61
Maldives 6.08
Mongolia 7.08
Myanmar 2.17
Nepal 5.05
Oman 3.12
Pakistan 4.35
Philippines 2.41
Qatar 0.26
Russian 5.01
Saudi Arabia 7.36
Singapore 3.62
South Korea 3.53
Sri Lanka 5.39
Syrian Arab Republic 10.57
 Tajikistan 7.75
Thailand 1.42
Timor-Leste 5.07
Turkey 13.39
Turkmenistan 5.08
United Arab Emirates 3.36
Uzbekistan 7.16
Vietnam 2.17
World 6.18
Yemen, Rep. 13.57

Table 5.1 List of Asian countries in the year 2021

There is only one type of model in our analysis, namely the Static Model. In this case, the Least
Square Method is applied, determining three equations: the linear equation, quadratic equation,
and single exponential equation. Microsoft Excel facilitated the computation of data for each
model in this project. For data forecasting, the relevant model equation was applied to meet the
study's objectives. The Root Mean Square Deviation (RMSD) was calculated for each model,
and the selection of the optimal model was based on achieving the lowest RMSD value.
5.1 STATIC MODEL
5.1.1 LINEAR EQUATION
The general form of Linear Equation is:
y = a0 + a1 x .

Where,

y = independent variable (year)

a0 , a1 = constant
x = dependent variable (Asian country)

Asian Country, x Year, y x2 xy

1 13.28 1 13.28
2 20.9 4 41.8
3 6.58 9 19.74
4 1.87 16 7.48
5 5.23 25 26.15
6 4.33 36 25.98
7 7.65 49 53.55
8 0.61 64 4.88
9 4.82 81 43.38
10 9.33 100 93.3
11 10.66 121 117.26
12 5.32 144 63.84
13 5.98 169 77.74
14 4.41 196 61.74
15 11.46 225 171.9
16 14.19 256 227.04
17 5.05 289 85.85
18 2.8 324 50.4
19 19.25 361 365.75
20 4.9 400 98
21 3.71 441 77.91
22 9.1 484 200.2
23 1.26 529 28.98
24 14.49 576 347.76
25 3.01 625 75.25
26 4.61 676 119.86
27 6.08 729 164.16
28 7.08 784 198.24
29 2.17 841 62.93
30 5.05 900 151.5
31 3.12 961 96.72
32 4.35 1024 139.2
33 2.41 1089 79.53
34 0.26 1156 8.84
35 5.01 1225 175.35
36 7.36 1296 264.96
37 3.62 1369 133.94
38 3.53 1444 134.14
39 5.39 1521 210.21
40 10.57 1600 422.8
41 7.75 1681 317.75
42 1.42 1764 59.64
43 5.07 1849 218.01
44 13.39 1936 589.16
45 5.08 2025 228.6
46 3.36 2116 154.56
47 7.16 2209 336.52
48 2.17 2304 104.16
49 6.18 2401 302.82
50 13.57 2500 678.5
  x = 1275  y = 325.95 x 2
= 42925  xy = 7731.26

From the data, we calculated a0 and a1 :

 n

 x  a  =   y  .
0

 x  x   a   xy 
2
1

Then substitute the value from the table calculated:

 50 1275   a0   325.95 
1275 42925  a  = 7731.26  .
  1  

Then we got the value of a0 and a1 which are:

a0 = 7.9405
.
a1 = −0.0557

Thus, the equation of a Linear is:

y = 7.9405 − 0.0557 x .

After that, we calculated the Predicted unemployment rate by using the equation obtained
above and the square of the Actual minus the Predicted as shown in the table below:

Actual ( Actual − Predicted )

2
Pr edicted (linear)
13.28 7.8848 29.10818304
20.9 7.8291 170.8484268
6.58 7.7734 1.42420356
1.87 7.7177 34.19559529
5.23 7.662 5.914624
4.33 7.6063 10.73414169
7.65 7.5506 0.00988036
0.61 7.4949 47.40184801
4.82 7.4392 6.86020864
9.33 7.3835 3.78886225
10.66 7.3278 11.10355684
5.32 7.2721 3.81069441
5.98 7.2164 1.52868496
4.41 7.1607 7.56635049
11.46 7.105 18.966025
14.19 7.0493 50.98959649
 5.05 6.9936 3.77758096
2.8 6.9379 17.12221641
19.25 6.8822 152.9624768
4.9 6.8265 3.71140225
3.71 6.7708 9.36849664
9.1 6.7151 5.68774801
1.26 6.6594 29.15352036
14.49 6.6037 62.19372769
3.01 6.548 12.517444
4.61 6.4923 3.54305329
6.08 6.4366 0.12716356
7.08 6.3809 0.48874081
2.17 6.3252 17.26568704
5.05 6.2695 1.48718025
3.12 6.2138 9.57159844
4.35 6.1581 3.26922561
2.41 6.1024 13.63381776
0.26 6.0467 33.48589689
5.01 5.991 0.962361
7.36 5.9353 2.02977009
3.62 5.8796 5.10579216
3.53 5.8239 5.26197721
5.39 5.7682 0.14303524
10.57 5.7125 23.59530625
7.75 5.6568 4.38148624
1.42 5.6011 17.48159721
5.07 5.5454 0.22600516
13.39 5.4897 62.41474009
5.08 5.434 0.125316
3.36 5.3783 4.07353489
7.16 5.3226 3.37603876
2.17 5.2669 9.59078961
6.18 5.2112 0.93857344
13.57 5.1555 70.80381025
 = 994.1279923
Then, calculate RMSD using formula:

 ( actual − predicted )
2

RMSD = i =1
m

Then,

994.1279923
RMSD =
50

= 4.4590 .
Hence, the RMSD of the Linear Equation is 4.4590 .
5.1.2 QUADRATIC EQUATION
The general form of Quadratic Equation is:

y = a0 + a1 x + a2 x2 .

Where,

y = independent variable (year)

a0 , a1 , a1 = constant
x = dependent variable (Asian country)

Asian Year, y x2 x3 x4 xy x2 y
country, x
1 13.28 1 1 1 13.28 13.28
2 20.9 4 8 16 41.8 83.6
3 6.58 9 27 81 19.74 59.22
4 1.87 16 64 256 7.48 29.92
5 5.23 25 125 625 26.15 130.75
6 4.33 36 216 1296 25.98 155.88
7 7.65 49 343 2401 53.55 374.85
8 0.61 64 512 4096 4.88 39.04
9 4.82 81 729 6561 43.38 390.42
10 9.33 100 1000 10000 93.3 933
11 10.66 121 1331 14641 117.26 1289.86
12 5.32 144 1728 20736 63.84 766.08
13 5.98 169 2197 28561 77.74 1010.62
14 4.41 196 2744 38416 61.74 864.36
15 11.46 225 3375 50625 171.9 2578.5
 16 14.19 256 4096 65536 227.04 3632.64
17 5.05 289 4913 83521 85.85 1459.45
18 2.8 324 5832 104976 50.4 907.2
19 19.25 361 6859 130321 365.75 6949.25
20 4.9 400 8000 160000 98 1960
21 3.71 441 9261 194481 77.91 1636.11
22 9.1 484 10648 234256 200.2 4404.4
23 1.26 529 12167 279841 28.98 666.54
24 14.49 576 13824 331776 347.76 8346.24
25 3.01 625 15625 390625 75.25 1881.25
26 4.61 676 17576 456976 119.86 3116.36
27 6.08 729 19683 531441 164.16 4432.32
28 7.08 784 21952 614656 198.24 5550.72
29 2.17 841 24389 707281 62.93 1824.97
30 5.05 900 27000 810000 151.5 4545
31 3.12 961 29791 923521 96.72 2998.32
32 4.35 1024 32768 1048576 139.2 4454.4
33 2.41 1089 35937 1185921 79.53 2624.49
34 0.26 1156 39304 1336336 8.84 300.56
35 5.01 1225 42875 1500625 175.35 6137.25
36 7.36 1296 46656 1679616 264.96 9538.56
37 3.62 1369 50653 1874161 133.94 4955.78
38 3.53 1444 54872 2085136 134.14 5097.32
39 5.39 1521 59319 2313441 210.21 8198.19
40 10.57 1600 64000 2560000 422.8 16912
41 7.75 1681 68921 2825761 317.75 13027.75
42 1.42 1764 74088 3111696 59.64 2504.88
43 5.07 1849 79507 3418801 218.01 9374.43
44 13.39 1936 85184 3748096 589.16 25923.04
45 5.08 2025 91125 4100625 228.6 10287
46 3.36 2116 97336 4477456 154.56 7109.76
47 7.16 2209 103823 4879681 336.52 15816.44
48 2.17 2304 110592 5308416 104.16 4999.68
49 6.18 2401 117649 5764801 302.82 14838.18
50 13.57 2500 125000 6250000 678.5 33925
 x = 1275  y = 325.95  x 2
=  x3 =  x4 =  xy =  x2 y =
42925 1625625 65666665 7731.26 259054.86
From the data, we calculated a0 and a1 ,

 n x x 2
  a0    y 
 3   
 x x x   a1  =   xy  .
2

 x 2  2 
 x x 3 4
  a2    x y 

Then substitute the value from the table calculated,

 50 1275 42925   a0   325.95 

 1275
 42925 1625625   a1  =  7731.26  .
 42925 1625625 65666665  a2  259054.86 

Then we got the value of a0 , a1 and a2 which are:

a0 = 10.1932
a1 = −0.3157 .
a2 = 0.0051

Thus, the equation of a Linear is:

y = 10.1932 − 0.3157 x + 0.0051x 2 .

After that, we calculated the Predicted unemployment rate by using the equation obtained
above and the square of the Actual minus the Predicted as shown in the table below:

Actual ( Actual − Predicted )

2
Pr edicted (quadratic)
13.28 9.8826 11.54232676
20.9 9.5822 128.0925968
6.58 9.292 7.354944
1.87 9.012 51.008164
5.23 8.7422 12.33554884
4.33 8.4826 17.24408676
7.65 8.2332 0.34012224
0.61 7.994 54.523456
4.82 7.765 8.673025
9.33 7.5462 3.18194244
10.66 7.3376 11.03834176
5.32 7.1392 3.30948864
5.98 6.951 0.942841
4.41 6.773 5.583769
11.46 6.6052 23.56908304
14.19 6.4476 59.94475776
 5.05 6.3002 1.56300004
2.8 6.163 11.309769
19.25 6.036 174.609796
4.9 5.9192 1.03876864
3.71 5.8126 4.42092676
9.1 5.7162 11.45010244
1.26 5.63 19.0969
14.49 5.554 79.852096
3.01 5.4882 6.14147524
4.61 5.4326 0.67667076
6.08 5.3872 0.47997184
7.08 5.352 2.985984
2.17 5.327 9.966649
5.05 5.3122 0.06874884
3.12 5.3076 4.78559376
4.35 5.3132 0.92775424
2.41 5.329 8.520561
0.26 5.355 25.959025
5.01 5.3912 0.14531344
7.36 5.4376 3.69562176
3.62 5.4942 3.51262564
3.53 5.561 4.124961
5.39 5.638 0.061504
10.57 5.7252 23.47208704
7.75 5.8226 3.71487076
1.42 5.9302 20.34190404
5.07 6.048 0.956484
13.39 6.176 52.041796
5.08 6.3142 1.52324964
3.36 6.4626 9.62612676
7.16 6.6212 0.29030544
2.17 6.79 21.3444
6.18 6.969 0.622521
13.57 7.1582 41.11117924
 = 949.1232364
Then, calculate RMSD using formula:

 ( actual − predicted )
2

RMSD = i =1
m

Then,

949.1232364
RMSD =
50

= 4.3569 .
Hence, the RMSD of the quadratic equation is 4.3569
5.1.3 SINGLE EXPONENTIAL EQUATION
The general form of single exponential equation is:

y = Ae Bx .

Where,

y = independent variable (year)

x = dependent variable (Asian country)
A = Initial value of y
B = Change in y with respect to x

From the general form, make a comparison with form to linear equations,

ln y = ln Ae Bx
ln y = ln A + ln e Bx
ln y = ln A + Bx

y = a0 + a1 x .

x y x2 xy ln y x ln y
1 13.28 1 13.28 2.586259 2.586259
2 20.9 4 41.8 3.039749 6.079498
3 6.58 9 19.74 1.884035 5.652104
4 1.87 16 7.48 0.625938 2.503754
5 5.23 25 26.15 1.654411 8.272056
6 4.33 36 25.98 1.465568 8.793405
7 7.65 49 53.55 2.034706 14.24294
8 0.61 64 4.88 -0.4943 -3.95437
 9 4.82 81 43.38 1.572774 14.15497
10 9.33 100 93.3 2.233235 22.33235
11 10.66 121 117.26 2.366498 26.03148
12 5.32 144 63.84 1.671473 20.05768
13 5.98 169 77.74 1.788421 23.24947
14 4.41 196 61.74 1.483875 20.77425
15 11.46 225 171.9 2.438863 36.58294
16 14.19 256 227.04 2.652537 42.4406
17 5.05 289 85.85 1.619388 27.5296
18 2.8 324 50.4 1.029619 18.53315
19 19.25 361 365.75 2.957511 56.19271
20 4.9 400 98 1.589235 31.7847
21 3.71 441 77.91 1.311032 27.53167
22 9.1 484 200.2 2.208274 48.58204
23 1.26 529 28.98 0.231112 5.31557
24 14.49 576 347.76 2.673459 64.16301
25 3.01 625 75.25 1.10194 27.5485
26 4.61 676 119.86 1.528228 39.73392
27 6.08 729 164.16 1.805005 48.73513
28 7.08 784 198.24 1.957274 54.80367
29 2.17 841 62.93 0.774727 22.46709
30 5.05 900 151.5 1.619388 48.58165
31 3.12 961 96.72 1.137833 35.27282
32 4.35 1024 139.2 1.470176 47.04563
33 2.41 1089 79.53 0.879627 29.02768
34 0.26 1156 8.84 -1.34707 -45.8005
35 5.01 1225 175.35 1.611436 56.40026
36 7.36 1296 264.96 1.99606 71.85816
37 3.62 1369 133.94 1.286474 47.59954
38 3.53 1444 134.14 1.261298 47.92932
39 5.39 1521 210.21 1.684545 65.69727
40 10.57 1600 422.8 2.35802 94.32079
41 7.75 1681 317.75 2.047693 83.95541
42 1.42 1764 59.64 0.350657 14.72759
43 5.07 1849 218.01 1.623341 69.80366
44 13.39 1936 589.16 2.594508 114.1584
45 5.08 2025 228.6 1.625311 73.13901
46 3.36 2116 154.56 1.211941 55.74928
47 7.16 2209 336.52 1.96851 92.51997
 48 2.17 2304 104.16 0.774727 37.1869
49 6.18 2401 302.82 1.821318 89.2446
50 13.57 2500 678.5 2.607861 130.3931
 x = 1275 y= x 2
=  xy =  ln y =  x ln y =
325.95 42925 7731.26 80.3745 1981.531

From the data, we calculated a0 and a1 ,

 n

 x  a  =   ln y  .
0

 x  x   a   x ln y 
2
1

Then substitute the value from the table calculated,

 50 1275   a0   80.3745 
1275 42925  a  = 1981.531 .
  1  

Then we got the value of a0 and a1 which are:

a0 = 1.774066531
.
a1 = −0.007

From the value a0 and a1 , comparing a0 and a1 with the linear equation and got the value of
A and B ,

InA = a0
B = a1
A = e1.774066531
B = −0.007
A = 5.8948

y = 5.8948e−0.007 x .

After that, we calculated the Predicted unemployment rate by using the equation obtained
above and the square of the Actual minus the Predicted as shown in the table below:

Actual ( Actual − Predicted )

2
Pr edicted (linear)
13.28 5.853680486 55.15022152
20.9 5.812847804 227.6221614
6.58 5.772299952 0.652379367
1.87 5.732034945 14.91531391
5.23 5.692050808 0.213490949
4.33 5.652345583 1.74859784
7.65 5.612917323 4.149705832
0.61 5.573764098 24.63895402
 4.82 5.534883989 0.511059117
9.33 5.49627509 14.69744669
10.66 5.457935509 27.06147497
5.32 5.419863368 0.009972692
5.98 5.382056802 0.357536068
4.41 5.344513958 0.873316337
11.46 5.307232996 37.85654181
14.19 5.270212089 79.56261638
5.05 5.233449424 0.033653691
2.8 5.196943199 5.745336698
19.25 5.160691625 198.5086105
4.9 5.124692926 0.050486911
3.71 5.088945338 1.901490245
9.1 5.053447109 16.3745903
1.26 5.018196501 14.12404094
14.49 4.983191785 90.37940244
3.01 4.948431246 3.757515696
4.61 4.913913181 0.092363222
6.08 4.8796359 1.440873973
7.08 4.845597721 4.992553544
2.17 4.811796978 6.979091271
5.05 4.778232013 0.073857839
3.12 4.744901183 2.640303855
4.35 4.711802854 0.130901305
2.41 4.678935404 5.148067869
0.26 4.646297223 19.23960333
5.01 4.613886712 0.156905737
7.36 4.581702282 7.71893821
3.62 4.549742356 0.864420849
3.53 4.518005369 0.976154609
5.39 4.486489765 0.816330746
10.57 4.455193999 37.39085243
7.75 4.424116539 11.0615008
1.42 4.393255862 8.840250419
5.07 4.362610455 0.500399969
13.39 4.332178817 82.04412459
5.08 4.301959456 0.605347088
3.36 4.271950892 0.83165443
7.16 4.242151655 8.513838964
2.17 4.212560284 4.172052514
6.18 4.18317533 3.987308765
13.57 4.153995351 88.66114354
 = 1118.774756
Next, calculate RMSD using formula:

 ( actual − predicted )
2

RMSD = i =1
m

Then,

1118.774756
RMSD =
50

= 4.7303 .
Hence, the RMSD of the single exponential equation is 4.7303 .
5.2 RESULT AND COMPARISON
Model Method RMSD
Static Model Linear Equation 4.4590
Quadratic Equation 4.3569
Single Exponential Equation 4.7303

As a comparison and result, based on the table above, it is concluded that the model
that perform the best forecasting is the Static Model by Quadratic equation. This is because the
RMSD value in Quadratic Equation is the smallest compared with the Linear Equation and
Single Exponential Equation.

6.0 CONCLUSION

In summary, this research underscores the utility of mathematical models for

comprehending and investigating the practical use of equations or functional connections.
Subsequently, gaining proficiency in mathematical modeling software, such as Microsoft
Excel, enhances our ability to implement these insights in our everyday experiences.
Consequently, through the application of learned concepts, we can deduce that among the
models, the Static Model within Quadratic Equations emerges as the most effective, as
evidenced by its minimal Root Mean Square Deviation (RMSD). Furthermore, employing the
identified equations enables us to forecast unemployment in Asian countries in 2021.

7.0 REFERENCES

Chamseddine, Z., & Boubkr, A. A. (2021). Understanding Gender, Income and Travel
Behavior in Casablanca City – Morocco. The Open Transportation Journal, 15(1),
272–279. https://doi.org/10.2174/1874447802115010272
Liu, H. (2007). Unemployment in Asia - Edited by John Benson and Ying Zhu. Economic
Record, 83(262), 349–351. https://doi.org/10.1111/j.1475-4932.2007.00422.x
MacRae, D. G. (2018). Thomas Malthus | Biography, Theory, Books, & Facts. In
Encyclopedia Britannica. https://www.britannica.com/biography/Thomas-Malthus
Naser, A. F., Mohammed, H. A., & Mohammed, A. A. (2021). Mathematical Modeling of
Linear Static and Dynamic Analysis for Pier Height Effect on the Structural
Performance of Bridges Structures. Mathematical Modelling of Engineering
Problems, 8(4), 617–625. https://doi.org/10.18280/mmep.080415
Pant, A. (2022). Unemployment dataset. Www.kaggle.com.
https://www.kaggle.com/datasets/pantanjali/unemployment-dataset
Russ, S. (2011). The unreasonable effectiveness of mathematics in the natural sciences.

London, Eng.

Sadiq, N., & Qureshi, M. S. (2010). Climatic Variability and Linear Trend Models for the
Five Major Cities of Pakistan. Journal of Geography and Geology, 2(1).
https://doi.org/10.5539/jgg.v2n1p83
Tejera-Vaquerizo, A., Cañueto, J., Toll, A., Santos-Juanes, J., Jaka, A., Ferrandiz, C.,
Sanmartín, O., Ribero, S., Moreno-Ramírez, D., Almazán, F., Fuente, M. J.,
Podlipnik, S., & Nagore, E. (2020). Estimated Effect of COVID-19 Lockdown on
Skin Tumor Size and Survival: An Exponential Growth Model. Actas Dermo-
Sifiliográficas (English Edition), 111(8), 629–638.
https://doi.org/10.1016/j.adengl.2020.09.008
The World Bank. (2023). Unemployment, total (% of total labor force) (modeled ILO
estimate). Worldbank.org. https://data.worldbank.org/indicator/sl.uem.totl.zs
TheGlobalEconomy. (n.d.). Unemployment rate in Asia. TheGlobalEconomy.com.
https://www.theglobaleconomy.com/rankings/unemployment_rate/Asia/
UN ESCAP. (2003). A note on unemployment in the wake of the Asian economic crisis and
some responses. Bulletin on Asia-Pacific Perspectives 2002/03: Asia-Pacific
Economies: Sustaining Growth amidst Uncertainties.
Verma, S. P., Andaverde, J., & Santoyo, E. (2006). Application of the error propagation
theory in estimates of static formation temperatures in geothermal and petroleum
boreholes. Energy Conversion and Management, 47(20), 3659–3671.
https://doi.org/10.1016/j.enconman.2006.02.028
Yuan, H., Li, Z., Wang, H., & Song, C. (2016). Static modeling and analysis of continuum
surgical robots. 2016 IEEE International Conference on Robotics and Biomimetics
(ROBIO 2016). https://doi.org/10.1109/robio.2016.7866333