How did Covid 19 Affect Strategic Goods? A Study using the Scenario

Method

SAWSSAN SAADAOUI1, MOHAMED BENMERIEM 2, HANANE ABDELLI1,

ZOUHEYR GHERAIA1

1Department of Business Management, Jouf University, Sakaka, SAUDI ARABIA

2Faculty of Economic Sciences, Commerce and Management Sciences, Chlef University, ALGERIA

Orcid ID Zouheyr Gheraia: https://orcid.org/0000-0002-8526-5205

Abstract: - Based on a review of literature dealing with the impact of the Covid-19 pandemic on international

transactions, it should be noted that this pandemic in the world has led to a radical change in several areas. In

this article, we have chosen to focus on the sectors that are strongly affected by this pandemic, namely the oil

sector and not forgetting to study fluctuations related to the metals sector (gold). Indeed, the oil sector is a

market that is well affected by this pandemic, which has caused significant price fluctuations. That is why this

study is trying to identify the impact of this pandemic on this sector. Similarly, due to the lack of stability in the

prices of metals (particularly gold), this leads us to analyze and process these fluctuations in order to determine

the effect caused by the pandemic.

Finally, to clarify our research questions, a study based on the scenario model is retained to identify the impact

of Covid-19 on each of these variables, at the beginning and to verify the effect between all these variables,

thereafter. The obtained results demonstrate that the pandemic affects negatively Oil prices. On the contrary,

the high number of infected people leads to the rise in gold price during the forecast period.

Key-Words: - Covid-19 pandemic, strategic goods, oil price, gold price, scenario method, Var model.

Received: August 21, 2021. Revised: June 5, 2022. Accepted: June 16, 2022. Published: July 8, 2022.

1 Introduction

Since the Second World War, the Covid 19

pandemic constitutes the global crisis and the

biggest challenge we have lived. Nevertheless, the

pandemic is much more than a health crisis. It is an

unprecedented crisis putting pressure on each

countries it affects.

COVID-19 pandemic is having a considerable

effect on all states around the world. It is

engendering the loss of many lives, affecting

people’s way of life and work, and bringing about

socio-economic changes that have a considerable

impact during the coming years. The measures taken

by States to deal with the pandemic are changing all

aspects of the economy and life. As a result, the oil

sector and the metals sector are under-going

dramatic changes.

Thus, according to the literature, many studies

tried to identify the impact of this pandemic in these

variables. At first, according to [10], the Covid 19

pandemic has an impact on the fluctuation of the oil

price. [9] who noted that the change in the oil price

remains to the Covid 19 outbreak approves this idea.

The prices of metals, especially, the precious metals

(gold) has seen a significant fluctuation and

movement during this pandemic [2], [7].

Therefore, to deal with these propositions, this

study aims to answer the following questions:

•Is there an effect of Covid 19 on oil price and the

precious metals (gold) price?

• Is there a relationship between the effects of the

Covid 19 on these variables?

2 Literature Review

Since the end of 2019, the world has been in a

critical situation following the appearance of the

COVID-19 pandemic. This pandemic has caused

many problems for countries including the strategic

changes that have affected the global economy and

international transactions. Beyond this crisis, the

Coronavirus pandemic triggered an economic crisis

that had dramatic effects on the financial markets.

An economic slowdown has been caused by the

COVID-19 pandemic that affects the oil demand

[1], [9], [10]. Not only the oil sector has been

affected by the pandemic but also the prices of

precious metals (especially gold) has seen a

significant fluctuation during the pandemic [2], [7].

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2022.18.90

Sawssan Saadaoui, Mohamed Benmeriem,

Hanane Abdelli, Zouheyr Gheraia

E-ISSN: 2224-3496

944

Volume 18, 2022

Starting with the oil sector, the COVID-19

pandemic has affected oil prices [2], [5]. An

imbalance has characterised the oil market in 2020.

The procedures taken by the while world to deal

with this pandemic has led to fluctuations in the

demand and supply of the oil market. Since March

2020, the oil market has undergone a radical change

in the oil prices. [10] states that an unprecedented

collapse in oil prices has been distinguished because

of this pandemic. The authors add that the crude oil

price fell by 85 percent from the date of 22 January

2020 when the first case of COVID-19 was

detected. It means that the crude oil price

experienced a sever fall of two-thirds since January

2020 to April 2020. According to a recent OPEC

report, it appears that in 2018 the price per barrel

was practicing the same as in 2017 (53.12 dollars) to

fall in 2019 up to 40.23 dollars the barrel. This price

will see an increase of up to 49.12 a barrel during

the year 2020. This is primarily related to the

current COVID 19 health crisis, which is causing

patterns energy consumption has changed. (OPEC,

2020, p145). Similarly, according to the World

Bank, as a result to this pandemic, the oil price fell

to 30 dollars in March and to 25 dollars in April

2020. Therefore, to summarize, the COVID-19

pandemic is negatively related to the oil prices,

which means that a negative relationship exists

between the oil prices and Qin et al. (2020) note the

COVID-19 pandemic as it.

Concerning the precious metals, we find that the

gold prices have increased 8 percent during

COVID-19 pandemic exactly from January 2020.

Similarly, [6] provide that gold prices has

experienced a slight fall at the beginning of the

crisis. Nevertheless, subsequently from February

2020, gold prices presented a considerable increase.

[3] indicate even with the pandemic the demand for

gold continues to increase, which causes a

continuous increase in its price. In addition, the

study presented by [11] provides that a pushing in

the price of the gold can be seen during the period

from 1-Jan to 9-Mar, from 1.517 dollar to 1.680

dollar. Although, a simple fall in the prices was seen

during March but this for a short period. Regarding

the relationship between oil prices and gold prices,

[4] indicated that the increase of gold prices is

related to crude oil prices. This relationship is linked

to the role of oil as a principal input for several

goods. In this way, it is important in this study to

demonstrate if the oil prices affect gold prices

during the COVID-19 pandemic.

3 The Standard Study (Knowing the

Behavior /Movement of the Internal

Variables during the Year 2022 by using

Scenarios)

3.1 Analyzing the Sensitivity between a Set of

Variables by using the Vector

Autoregression (VAR):

In order to recognize how both oil and gold prices

are sensitive and closely related in the context of the

coronavirus disease 2019 (COVID-19), we estimate

the VAR model to highlight the relationship

between variables in the price of oil, the price of

gold and the number of people infected by the

pandemic in the world.

The aim is to determine the direction in which

the two variables will move, especially when

changes and innovations occur in the number of

infected people, taking into consideration, the

renewed global changes. This can be done through,

on the one hand, the analysis of both response

functions variance and, on the other hand, analyzing

the possible scenarios provided by the VAR model

between the variables in order to know the size and

direction of the change of these two variables by

causing any value change in the global number of

infected people.

However, before considering and estimating the

VAR model, the following important steps must be

taken into consideration:

3.2 Studying the Stability of Study Variables

Based on the evolution curve of the study variables

during the period of study, it appears that the series

initially contain a general trend, which firstly

suggests the instability of the time series. In order to

do this, the stability of the study variables is tested

by using the developed Dickey Fuller test

and Philips Peron (PP). However, before that, since

we include monthly-frequency variables, it is

necessary to remove the Compounding Quarterly

Formula (in case it exists) before the test that cares

about the general trend.

With regard to the results, which includes parameter

values of the seasonal component and the partial

correlation coefficient (correlogram) that we got

directly through EViews program 12.0, it is worth

noting that the variables are free from the Quarterly

Compounding, where the coefficients values were

almost closer to 1. Therefore, there is no significant

difference between the sample variables and the

variables that are compounded by the seasonality

coefficients. It shows uniform rotundity in the

autocorrelation coefficient during the study period.

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2022.18.90

Sawssan Saadaoui, Mohamed Benmeriem,

Hanane Abdelli, Zouheyr Gheraia

E-ISSN: 2224-3496

945

Volume 18, 2022

Relying on the Stability Testing software prepared

in the Eviews 12.0²² program for ADF test, we

obtained the following results in the following

Figure.

Variables Augmented Dickey–

Fuller test results Phillips–Perron test

results

ADF

with

trend

ADF

without

trend

PP with

trend

without

trend

Level

1st Difference

Level

1st Difference

Level

1st Difference

Level

1st Difference

NCM

-2.934

-6.139***

-2.899

-6.064***

-2.156

-6.227***

-3.125

-6.122***

-4.764***

-4.709***

-3.257**

-3.212*

COP

*,**,*** Show that the variables are significant at1%, 5% and 10% levels,

respectively.

Fig. 1: ADF Test results for the Stationality

(stability) in series

Source: Authors

The present Fig. 1 shows us that, concerning the

two series

NCM

, the values of the calculated

statistics related to ADF and PP tests are

smaller (in absolute value) than the tabulated

statistics in the three models at the 5% level

of significance, that’s why we accept the hypothesis

Also, the non- significance of the general trend in

the third model for both series shows that the two

series are of type DS. Nevertheless, after the first

difference is made, he calculated statistics values

become greater (in absolute value) than the

tabulated statistics in the three models at the

level of significance 5%. Hence, the two series

NCM

are stationary (stable), whereas the

series

COP

of type TS are unstable and can be

made stationary by using regression. Whereas the

equation of a trend line that is subtracted from the

original series to become stable takes the following

form:

tPOC t⋅+= 13,231,39

where we get the

stable series

COPS

. In addition, the instability

resulting from the presence of a significant

structural change in all series, has been tested by

knowing the significance of the coefficient of the

endemic variable that represents the time of

occurrence of the structural change in the series.

Therefore, based on Eviews 12.0 program, we note

that the probabilities corresponding to the statistic

of the two parameters BREAKDUM in the two

series

NCM

COP

that are

( )

05,017,0 〉=

cncm

( )

05,042,0 〉=

cgp

( )

05,037

,0 〉=

ccop

proves that the chains are free

from the problem of structural change, which may

lead to its instability.

3.3 Test of the Co-integration Relationship

Until now, we found that the two series

COP

are unstable, but they are not homogeneous in terms

of instability as the series

is of type DS but the

series

COP

is of type TS. Therefore, we can say

that there is no room for joint integration (a long-

term relationship) between these two variables

according to the conditions of the Cointegration

test .Also, given that these two variables represent

the internal variables of the VAR model that we

want to estimate, there is no need to use the VECM

error correction model in the estimation.

3.4 Estimated Results of VAR Model

The VAR model is successfully applied on

identifying the size and nature of the relationship

between the variables

NCM

COP

However, before doing this, it is necessary to

determine the optimum delay. So, with regard to

figure 02, we note that the optimal degree of delay

according to most of the statistical criteria is

2=p

(Based on Eviews 12.0 program).

tabulè

( )

tabulè

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2022.18.90

Sawssan Saadaoui, Mohamed Benmeriem,

Hanane Abdelli, Zouheyr Gheraia

E-ISSN: 2224-3496

946

Volume 18, 2022

Fig. 2: Optimum Delay degree, Source: Authors

In addition, one of the conditions for estimating the

VAR model is using stable series. We can

summarize the estimated results of VAR model

according to the following equation after removing

the insignificant variables at the 10% level of

significance1.

Estimated results of oil price model

: BN * Significant at 10%، ** Significant at 5%،

*** Significant at 1%، Source: Authors

The results of the estimation can be analysed as

follows:

• There is a proportional relationship between

the oil price a month ago and now.

Therefore, if the price of oil rose by 1

dollar last month this would also lead to

an increase by $ 0.73 in today’s price.

• There is an inverse relationship between the

oil price of two month-late period and now.

In that, if the oil price rose by $1, this would

lead to a decrease by $ 0.45 in today’s price.

• There is a significant and inverse

relationship between the gold price of two-

month late period and the current price.

Therefore, if the price of gold rose by $1,

this would lead to a decrease in today’s

price by $0.03. Because oil and gold

constitute two alternative articles, so when

1 We can find the Fisher and Student values by using eviews

program @qtdist(0.95,20)=1,66 ،@qtdist(0.90,20)=1,29 ،

2,37@qtdist(0.99,20)=،=2,58@qfdist(0.90,2,20)

demand of one increases, the other’s price

decrease.

• There is an inverse relationship between the

numbers of people infected by the

coronavirus disease in one or two- month

late period and the price of oil since an

increase of 1 million in the number of

infected people would lead to today’s oil

prices drop by-0.46 and -0.45 dollars,

respectively. In this way, the increase

spread of the epidemic leads to increased

closure that means a significant drop in

demand.

• In general, the fluctuation of the oil price is

related to its dynamism based on its current

price that depends on the past price of oil.

Therefore, this proportional relationship is

due to the high demand for the oil in the

short term. In addition, on the one hand, the

transactions of oil sales, are forward

contracts with relatively long period. On the

other hand, there is an inverse relationship

between the current price of oil and the two

periods of delay, which makes economic

sense. In this way, the rise of the oil price

leads to the lower demand, which in turn

reduces the price to increase demand.

Therefore, generally speaking, economic

interpretations of oil prices are limited due

to its attachment to the international

policies, security, and based on cartel

decisions such as OPEC.

The results of Gold price estimation

The results of the estimation can be analysed as

follows :

• There is a proportional relationship between

the price of gold last month and the current

price .If the gold price rose by 1 dollar the

last month , this would lead to an increase in

the current price of gold by 0.47 dollars.

This result leads to a reduction in the

counties’ imports, their trade balance and

the overall product cost. It leads, also, to an

improvement in country’s crude domestic

output, which increases national income and

increases the demand for gold as a strategic

commodity. In this regards, there is a rise in

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2022.18.90

Sawssan Saadaoui, Mohamed Benmeriem,

Hanane Abdelli, Zouheyr Gheraia

E-ISSN: 2224-3496

947

Volume 18, 2022

the price of gold by producing countries due

to the increased demand and urgency by

most countries.

• There is an inverse relationship between the

number of people who had been infected by

the coronavirus disease in one month late

period and the price of gold. In fact, the

one-million infected people will lead to an

increase in the current gold price by $5.04.

In this way, the rise in the number of people

infected by the epidemic forces countries to

pursue a policy of total closure, resulting in

a sharp decline in domestic and global

demand for most goods and a halt in the

production of most enterprises. This will

contribute to the decline in dollar prices,

which will lead to a rise in the price of gold

as an alternative commodity to the price of

the dollar.

4 Model Validation

Many tests should be done to verify the validity of

the results of the var (2) model and, most

importantly, to demonstrate that all the residuals

resulting from the model are Noise trade (stable)2.

This is confirmed by the partial and autocorrelation

functions of the residual series. Therefore, by

looking at the graphic representation of the partial

autocorrelation function for the residuals of the var

model, we notice that all the simple and partial

autocorrelation coefficients are located inside the

Confidence Interval. This indicates that these

Correlation Coefficients do not differ from zero at

the 5% level of significance. Therefore, the residual

walkways of the var model can be considered as

white noise. It can be confirmed by the Ljung-Box

test, where we find that A significant P-value in The

Ljung-Box Test statistic is and

respectively, which completely

greater than 5%. Therefore, we accept the null

hypothesis, that is, the residual walkways of the Var

model is white noise and therefore these residual

series can be considered as stable.

2 If the residuals are white noise (noise trade), that is, they are

stable and this will avoid the most important problems of

estimation. There is no problem of correlation of residuals of

the first degree or degree p, there is no problem of heterogeneity

of residual variance

4.1 Testing the Stationary VAR Model

(Testing the Stability of the VAR Model)

We can say that the radial path with

dimension (n, 1) subject to a representation from the

form VAR (p) is stable. If all the values of the roots

of the polynomial Inverse time delay parameters are

less than one in absolute value, by using the 12.0

Eviews program directly, we get the

following figure.

Fig. 3: VAR model stability conditions, Source:

Authors

This figure shows that the estimated model fulfills

the requirements for stability as all coefficients are

less than one, and all roots are located within the

unit circle, which indicates the structural stability of

the Var model as a whole and that this model is

valid for forecasting.

4.2 Scenario Analysis: Forecasting the

Future Values between COVID-19

Pandemic, Oil Price and COVID-19 Deaths

Cases shocks

The main advantage of these scenarios that are

extracted from the results of the previous Var (2)

model and obtained directly using Eviews 12.0

program, is to know the size and direction of the

impact of any change in any internal variable of the

VAR model on the future values of the same

variable and other variables in the model. However,

in the Shock Analysis, the results are obtained

automatically, and the standard deviation value of

the variable is often taken as a percentage change in

its value without changing this shock size.

( )

3,0=LBprob

( )

51,0=LBprob

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2022.18.90

Sawssan Saadaoui, Mohamed Benmeriem,

Hanane Abdelli, Zouheyr Gheraia

E-ISSN: 2224-3496

948

Volume 18, 2022

It is of a great importance now to present

forecasting results by using different scenarios.

However, before that, we need to predict and

forecast an exogenous3 variable

DNCM

according to the Box-Jenkins method for ARIMA

linear time series models, where we find that the

optimal model for this series

( )

( ) ( )

)2(66,071,00,2

78,363,1

ARARMADNCM

⋅−=→

−

, is

where the series

DNCMF

, that represents the

forecasted values until the end of 2022, is used in

the events of different scenarios.

An important first step is to forecast the future

values of the internal variables (the baseline) based

on the outcomes and results of the previous Var (2)

model.

The results of this prediction are called Baseline and

in order to make this baseline prediction in Eviews

12.0 we should:

• First, we extend the sample size until the

period in which we want to predict its future

values. We want to predict the future values

of the price of gold and oil for the next year

(until 12/2022).

• Then, The VAR (2) model should be re-

estimated.

• After that, based on the results of the Var

model, we create a simple prediction by

developing a model, which includes the

prediction period (from 12/2021 to

12/2022).

• After activating it, its formula is as follows :

Model: Untitled

Date: 02/26/22 Time: 18:12

Sample: 2021M12 2022M12

Solve Options:

Dynamic-Deterministic Simulation

Solver: Broyden

Max iterations = 5000, Convergence = 1e-08

Parsing Analytic Jacobian:

0 derivatives kept, 0 derivatives discarded

Scenario: Baseline

Solve begin 18:12:39

Solve complete 18:12:39

3 The prediction of the future values of the internal variables

(the baseline) is done according to the Box-Jenkis method for

ARIMA linear time series models, where we find that the VAR

models are nothing but a generalization of the latter, and there

are two types of this simple initial prediction: dynamic

(DYNAMYQUE SOLUTION) as in Our case, which takes past

values into account while making a forecast, and STATIC

SOLUTION, which depends only on current values in

forecasting

EVIEWS 12.0 allows us to obtain the predictive

values (baseline) shown in the figure 4.

Fig. 4 : The baseline predictive values of monthly

prices for gold and oil from December2021 to

December 2022, Source : Authors

Generally speaking, we notice that the predictive

values during the relevant period (from 12/2021 to

12/2022) with an average of -2,96 and 14.16 for oil

and gold prices are stable and these are the simple

predictive values (baseline).

However, during the forecast period, the values of

the oil price are all negative because the predictive

values are less than zero whereas the values of the

gold price is positive (greater than0).

Oil prices went negative due to the effect of the

Coronavirus pandemic worldwide. On the contrary,

the rise in gold price during the forecast period can

be attributed to the high number of infected people.

Scenario 1

A scenario can be created by making a specific

change in the number of infected people to see its

impact on the predictive values of the two internal

variables (oil and gold price). For example, if we

suppose that the number of infected people

increased by 30% (which represents the average

growth during the study period) during the forecast

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2022.18.90

Sawssan Saadaoui, Mohamed Benmeriem,

Hanane Abdelli, Zouheyr Gheraia

E-ISSN: 2224-3496

949

Volume 18, 2022

period. By using the Eviews output, we can obtain

the following results.

Fig. 5: Predictive values for the price of gold and oil

as a result of the increase (30%) in the number of

monthly-infected people. Scenario Results (Scenario

1), Source: Authors

We note that there is an increase of 30% in the

number of people with coronavirus over the

predictive values (i.e. the baseline), which will lead

to a slight decrease in the monthly oil price (-3.45

dollars on average), but a small increase in the price

of gold during the forecast period.

Scenario 2 : This will be totally different from the

previous one. In that, it is possible to reduce the

number of infected people by 30% (which

represents an average growth during the study

period) to see its impact on the predictive values of

the two internal variables (oil and gold price).

During the same period (from 12/2021 to 12/2022)

and by using the outputs of eviews, the results can

be as follows.

Fig. 6 : Predictive values for the price of gold and

oil as a result of the decrease (30%) in the number

of monthly infected people. Scenario Results

(Scenario 2), Source: Authors

It is worth mentioning that, on the one hand, the

decrease of 30% in the number of infected people

from 12/2021 to 12/2022 will to a slight increase in

the price of oil (-2, 94), which is slightly greater

than the forecast standards. On the other hand, the

decrease in the number of infected people will lead

to a decrease in the price of gold (-2.94 on average)

during the relevant period.

5 Conclusion

In this paper, we investigate the relationship

between the price of gold and oil during the Covid-

19 pandemic. We find that, in general, the decrease

or changes in the price of gold and oil when the

number of infected people changes in the different

scenarios can be explained to the form of the

estimated Var model, as most of the variables are

not explanatory. In this regards and according to our

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2022.18.90

Sawssan Saadaoui, Mohamed Benmeriem,

Hanane Abdelli, Zouheyr Gheraia

E-ISSN: 2224-3496

950

Volume 18, 2022

results, it is clear that Oil prices went negative due

to the effect of the Coronavirus pandemic

worldwide. On the contrary, the rise in gold price

during the forecast period can be attributed to the

high number of infected people.

Acknowledgements:

This work was funded by the Deanship of Scientific

Research at Jouf University under grant No (DSR-

2021-04-0314)

References:

[1] Devpura, N. and Narayan, P.K. (2020),

Hourly oil price volatility: the role of COVID-

19, Energy Research Letters, Vol. 1, No. 2,

pp. 1-5, doi 10.46557/001c.13683.

[2] Farid, S.; Kayani, G. M.; Naeem, M. A.;

Shahzad, S. J. H. (2021), Intraday volatility

transmission among precious metals, energy

and stocks during the COVID-19 pandemic ,

Resources Policy ; Vol. 72, August 2021.

[3] Grima, S., Dalli Gonzi, R., Thalassinos, I.E.

(2020), the Impact of COVID-19 on Malta

and its Economy and Sustainable Strategies,

Available at SSRN:

https://ssrn.com/abstract=3644833 or

http://dx.doi.org/10.2139/ssrn.3644833.

[4] Hooker, M.A. (2002), Are oil shocks

inflationary? Asymmetric and nonlinear

specifications versus changes in regime,

Journal of Money, Credit and Banking, Vol.

34.

[5] Kartal, M.T., Depren, Ö., & Kılıç Depren, S.

(2020), The Determinants of Main Stock

Exchange Index Changes in Emerging

Countries: Evidence from Turkey in Covid-19

Pandemic Age, Quantitative Finance and

Economics, Vol.4, No.4, pp. 526–541.

https://doi.or g/10.3934/qfe.2020025.

[6] Mensi W, et al. (2020), Impact of COVID-19

outbreak on asymmetric multifractality of

gold and oil prices, Res Policy. 2020;

69:101829. Doi: 10.1016/j.reso] [CrossRef]

[Google Scholar] urpol.101829. [PMC free

article]

[7] Mokni, Khaled & Al-Shboul, Mohammed &

Assaf, Ata, (2021), Economic policy

uncertainty and dynamic spillover among

precious metals under market conditions:

Does COVID-19 have any effects?, Resources

Policy, Elsevier, Vol. 74, No, C.

[8] Qin, M., Zhang, Y.C., & Su, C.W. (2020), the

Essential Role of Pandemics: A Fresh Insight

into the Oil Market, Energy Research Letters,

Vol. 1, No. 1, 13166.

https://doi.org/10.46557/001c.13166.

[9] Shaikh, I. (2021), on the relation between the

crude oil market and pandemic Covid-19,

European Journal of Management and

Business Economics, Vol. 30, No. 3, pp. 331-

356. https://doi.org/10.1108/EJMBE-08-2020-

0223 Download as .RIS

[10] Wheeler CM, Baffes J, Kabundi A, Kindberg-

Hanlon G, Nagle PS, Ohnsorge F. (2020),

Adding fuel to the fire : cheap oil during the

COVID-19 pandemic, In : Prospects Group

Analysis. Washington, DC: The World Bank

Group. 10.1596/1813-9450-9320 [CrossRef]

[Google Scholar].

[11] Yousef, I, Shehadeh, E. (2020), the Impact of

COVID-19 on Gold Price Volatility,

International Journal of Economics &

Business Administration (IJEBA), Vol. VIII,

No. 4, pp. 353-364.

Creative Commons Attribution License 4.0

(Attribution 4.0 International, CC BY 4.0)

This article is published under the terms of the

Creative Commons Attribution License 4.0

https://creativecommons.org/licenses/by/4.0/deed.en

_US

Annex: Vector autoregression (VAR)

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2022.18.90

Sawssan Saadaoui, Mohamed Benmeriem,

Hanane Abdelli, Zouheyr Gheraia

E-ISSN: 2224-3496

951

Volume 18, 2022