Economic Growth Analysis Report: Principles of Data Science Study

This report explores the factors contributing to economic growth using data science principles, analyzing a dataset from 1960 to 2015 with 144 factors, including trade in services, exports, and agricultural imports. It details data workflows and dashboard design using Tableau, employing geo maps for visualization and regression analysis for prediction. ANOVA is used to determine the impact of different factors, and the analysis reveals insights into the economic growth of various countries and income groups. The report emphasizes reducing tariff barriers to improve trade and highlights the shift towards data-driven operations through accessible graphical representations, concluding that data visualization is crucial for understanding and improving economic scenarios.

Running head: PRINCIPLES OF DATA SCIENCE
Principles of Data Science
Name of the Student
Name of the university
Authors note

Introduction
Economic growth is considered as a process that includes different factors and helps
in improving the quality of the life of people. The selected data set contains multiple factors
that contributes to the economic growth of the countries. For the chosen dataset, the data
contains data from the year 1960 to 2015. In contains total 144 factors that mainly
contribute to the growth of different countries which includes “% of commercial service
imports”, trade GDP (as % of GDP), performance index for logistics and values for different
export and import factors related to the different countries.
The following report contributes to the process of exploration of chosen data set,
detailed explanation about the data workflows and dashboard design process. In addition to
that, critical evaluation of the approaches for analysis of the dataset.
Exploration of Dataset
In the given dataset, there are total 38018 rows of that contains all the data for the
countries and their respective values for parameters. Due to the large number of variables in
the dataset, it is reshaped in order to make it cleaned and optimized for the analysis process.
The selected parameters for this analysis are, Trade in services (% of GDP), Exports
of goods and services (BoP, current US$), Agricultural raw materials import (% of
merchandise imports) from the different countries.
Data Workflows & Proposal for Design of Dashboard
For the sheet1, sheet 2 and sheet 3 we have used the geo map visual representation.
The reason behind this can be stated through the use of this visualization technique it
becomes easy to understand the distribution values in the different countries as available in
the dataset. Moreover, the Easy to represent higher amount of values when compared to the

tabular or any other graphical representation of the large dataset. In this way, the visualization
helps in more intuitive decision making for the analyst. In the first tableau sheet the Max for
the GDP factors is used in order to visualize the gap for the same values for different
countries.
In the other sheets, the visualization includes the Export of Goods and services (%
GDP) and agriculture raw material is considered for the prediction of different countries.
Growth in exports is helpful in generating employment. For example, the growth in
manufacturing exports leads to creation of numerous jobs in the concerned industries, such as
car factories or other merchandise/agriculture.
In the traditional way the export jobs are mainly related to the manufacturing
industries. This is considered as an important source of full-time employment for improved
economic growth in the industrial regions.

Analysis and Critical Evaluation of the approaches
For the prediction of Economic growth, a variable is created named Predicted_GDP2,
which is calculated using the following script;
Predicted_GDP2
Role:
Continuous Measure
Type:
Calculated Field
Status:
Valid
Formula

SCRIPT_REAL("
x <- .arg1;
y <- .arg2;
z <- .arg3;
linreg <- lm(x ~ y +z)
linreg$fitted
", SUM([GDP]),SUM([Agri]),SUM([Trade])
)
In order to determine the impact of different factors on the economic growth of the
different countries the ANOVA technique is used. Analysis of variance or the ANOVA is
used in order to find out if mean of the two or more selected data groups significantly
different when compared to each other. The ANOVA technique helpful in finding the impact
of a single or more than two factors by relating the means of the different samples that are
considered.
For the comparison of export of agricultural raw materials exports of different
countries, the following dashboard is created.

The next prediction is done for the different countries of regions of the world. For
this predictions the regression technique is used. Using this technique mean of the selected
dependent variable are measured from the given specific values in the data sheet.
From the above dash board, it is evident that the countries in the Latin region has the
maximum predicted economic growth among all the other countries.

From the analysis it is important reduce tariff barriers business. Lower tariff barriers
are helpful in increasing trade for the countries. However, if the countries reduce general
tariff barriers, some domestic industries potentially lose business as they can no compete.
Reducing non-tariff barriers for business. Modern trade rules stress the importance of
non-tariff barriers and difficulties to trade. Eliminating these can help make trade more easy
and improve exports. This makes EU region based countries a single market may increase
the issues in trading new non-tariff barriers to trade.
Another, dash board shows the growth of the different income groups which is
depicted below;
From the above tableau dashboard, it can be stated that there is growth for the high
income group people depending on the considered factors. For the prediction, p-values of the
test indicated that values of both the linear and squared terms in the result are statistically
significant and differs from each other. Based on the available information from the test the

prediction model is created which provided the unbiased and statistically significant fit to the
available data and got the above prediction model.
Conclusion
It can be stated that visualization of the insights is no longer restricted to specialized
applications and specialized persons. Using the data visualization in different forms such as
maps, charts, as well as other graphical representations enables the users to better understand
available data to achieve tactical and strategic objectives for improving the scenario and
growth of the country or the region. In the regression analysis, range of prediction interval is
kept wider compared to the confidence interval. As due to the inclusion of multiple null
values the prediction includes greater uncertainty in the prediction process for an individual
depending on the selected factors rather than the mean of the factors. In this way, it is
possible to make the predictions as accurate as possible.
Furthermore, data visualization is stimulating a cultural shift toward added analytic,
data-driven operations and approaches by authorizing the end users to explore, in a graphical
manner which was previously accessible only in tabular report formats.