ANNALS OF THE “CONSTANTIN BRÂNCUȘI” UNIVERSITY OF TÂRGU JIU
LETTER AND SOCIAL SCIENCE SERIES
ISSN-P: 1844-6051 ~ ISSN-E: 2344-3677
https://alss.utgjiu.ro
2/2023
PREDICTIVE POWER OF ASYMMETRIC GARCH MODELS IN
VOLATILITY ESTIMATION: A CASE STUDY FOR
SWITZERLAND STOCK EXCHANGE
Puja KUMARI
Department of Commerce and Business Management, Ranchi University, Ranchi, India
Sunil KUMAR
Department of Economics, Purnea College, Purnia, Bihar, India
Ramona BIRAU
Faculty of Economic Science, University Constantin Brancusi of Tg-Jiu, Romania
Cristi SPULBAR
University of Craiova, Faculty of Economics and Business Administration, Craiova, Romania
Bharat Kumar MEHER
P.G. Department of Commerce & Management, Purnea University, Purnea, Bihar, India
Mukesh PASWAN
P.G. Department of Commerce & Management, Purnea University, Purnea, Bihar, India
Gabriela Ana Maria LUPU (FILIP)
University of Craiova, Doctoral School of Economic Sciences, Craiova, Romania
Abstract:
IN THE STOCK MARKET, VOLATILITY IS A TERM USED TO DESCRIBE THE
DEGREE TO WHICH THE PRICES OF ASSETS FLUCTUATE AND DETERMINES
THE DEGREE OF RISK OR UNCERTAINTY. THE MAIN AIM OF THE PRESENT
STUDY IS TO MODELING THE BEHAVIOR OF THE SWITZERLAND STOCK
MARKET USING DATA FROM 4TH JANUARY, 2000 TO 9TH NOVEMBER, 2023.
THROUGH THE APPLICATION OF GARCH FAMILY MODELS WHICH, INCLUDE
GARCH/TARCH, EGARCH, COMPONENT ARCH (1,1), AND PARCH. THE STUDY
USED A SAMPLE NUMBER OF 5994 DAILY OBSERVATIONS FOR SWISS STOCK
INDEX REPRESENTING THE SWITZERLAND STOCK MARKET. WE USED SOME
STATISTICAL TECHNIQUES SUCH AS PHILLIPS-PERRON AND AUGMENTED
DICKEY FULLER TESTS STATISTIC. THE ARCH LAGRANGE MULTIPLIER (LM)
TEST, PARCH MODEL. WE UTILIZED THE EVIEWS 12 ECONOMETRICS
PACKAGE. THIS STUDY HIGHLIGHTS THE SIGNIFICANCE OF ACCURATELY
AND METICULOUSLY SIMULATING STOCK MARKET BEHAVIOR IN ADDITION
73
ANNALS OF THE “CONSTANTIN BRÂNCUȘI” UNIVERSITY OF TÂRGU JIU
LETTER AND SOCIAL SCIENCE SERIES
ISSN-P: 1844-6051 ~ ISSN-E: 2344-3677
https://alss.utgjiu.ro
2/2023
TO ADDING TO THE CORPUS OF KNOWLEDGE IN FINANCIAL
ECONOMETRICS. THE CONCLUSIONS AND METHODS DISCUSSED IN THIS
STUDY PROVIDE A STRONG BASIS FOR FURTHER RESEARCH, ENHANCING
OUR CAPACITY TO PREDICT MARKET MOVEMENTS AND MAKE WISE CHOICES
IN A VOLATILE FINANCIAL ENVIRONMENT.
Keywords:
Contact details
of the
author(s):
VOLATILITY CLUSTERS, SWITZERLAND STOCK MARKET, FORECASTING,
GARCH FAMILY MODELS
.
Email: pujakumari254678@gmail.com
sunil197779@gmail.com
ramona.f.birau@gmail.com
cristi_spulbar@yahoo.com
bharatraja008@gmail.com
mukeshpaswan19850309@gmail.com
Lupuanamariagabriela@yahoo.com
Introduction
In the stock market, volatility is a term used to describe the degree to which the prices of assets (which
are seen as unpredictable) fluctuate and determines the degree of risk or uncertainty (Kim & Won,
2018). It swiftly causes prices to rise and fall (Badarla et al., 2021). A fortune may be made or lost in
the blink of an eye in the complex and volatile world of global finance. In this erratic environment, the
concept of stock market volatility is a critical factor in determining investment choices, risk
and mitigation tactics. It is crucial to quantify and examine volatility in the financial markets in order
to appropriately mitigate against instability. It's also beneficial to comprehend the level of risk
associated with any specific market (Badarla et al., 2021). In the financial markets, volatility is
important for hedging techniques, portfolio risk management, and derivative pricing.
As a result, accurate volatility prediction is paramount (Kim & Won, 2018). Since (Bollerslev, 1986)
introduced model which is ARCH models, it is a statistical model used to analyze volatility in time
series in order to forecast future volatility. Under the assumption that there is a probability density
function of the returns series, the parameters of the ARCH class models are generally estimated using
the parametric estimation method (Sun & Yu, 2019). Due to the capacity to capture volatility
persistence or clustering, the ARCH class models are advantageous (Bollerslev, 1986; Vedat, 1989;
Baillie, Bollerslev, & Mikkelsen, 1996). However, some existing studies have indicated that the ARCH
class models must be transformed to provide good forecasting performance (Choudhry & Wu, 2008).
In recent years, researchers have combined the GARCH model and computational intelligence-based
techniques for financial time series forecasting. (Engle, 1982) suggested the GARCH model, it is a
statistical model used in analyzing time-series data where the variance error is believed to be serially
autocorrelated.
The Swiss Market Index (SMI) is Switzerland's blue-chip stock market index, which makes it the most
followed in the country. It is made up of 20 of the largest and most liquid Swiss Performance
Index (SPI) stocks. As a price index, the SMI is not adjusted for dividends. The present effort focuses
on modeling the behavior of the Switzerland stock market using data from 4th January, 2000 to 9th
November, 2023. Insightful details on historical trends and volatility are provided by the detail analysis
74
ANNALS OF THE “CONSTANTIN BRÂNCUȘI” UNIVERSITY OF TÂRGU JIU
LETTER AND SOCIAL SCIENCE SERIES
ISSN-P: 1844-6051 ~ ISSN-E: 2344-3677
https://alss.utgjiu.ro
2/2023
for a particular market, supporting any future decision-making. We propose generalized autoregressive
conditional heteroscedasticity (GARCH) model to forecast stock price volatility in Switzerland stock
market.
Review of literature
Numerous research projects have been conducted to modeling the volatility of stock markets with the
help of various indexes. Birau et al. (2021) have studied the GARCH-based model behavior of the
stock markets in Hong Kong and Spain. In addition, (Spulbar et al., 2020) examined the Hong Kong
stock market's dynamics using short-term momentum effects. (Meher et al., 2020) investigated the
fluctuations in the market during the COVID-19 outbreak. However, the adverse information has far
stronger ramifications. Sokpo et al. (2017) conducted a study and found that the model series had
strong persistence, indicating that the market will be affected for some time by a positive or negative
shock to the stock market return series caused on by either good or bad news. (Spulbar et al., 2023)
argued that it is evident from the detrimental impacts of the global financial crisis that investors were
not able to make any significant profits from the Poland stock market. Furthermore, adverse shocks
occur more frequently than favorable ones. Bonga (2019) concluded that there is a positive
association between returns, risks, and volatility. As market volatility increases, the financial market
becomes more volatile.
Several research conducted worldwide have examined the volatility patterns and behavior of the stock
market using the GARCH family model such as Spulbar et al. (2022). Kumar et al. (2023b) conducted
a research study where the S&P Toronto stock index's volatility was examined using the EGARCH,
TGARCH, MGARCH, and PGARCH models. According to the paper's conclusion, the GARCH-GJR
model is better suitable. Kumar et al. (2023c) elaborated an empirical study that used the GARCH
(1,1), GJR-GARCH, EGARCH, M GARCH, and TGARCH models to assess the volatility of the
IBOVESPA index from stock market in Brazil. Aside from that, this study used both univariate and
multivariate models to assess the accuracy of volatility projections. Moreover, Maqsood et al. (2017)
conducted a research study that employed GARCH-M (1,1), EGARCH (1,1), TGRACH (1,1), and
PGARCH (1,1) models to measure the Nairobi Securities Exchange's volatility.
Out of different symmetric and asymmetric type heteroscedastic processes, they concluded that the
TGARCH (1,1) model is more suited to capture the volatility clustering and leverage impact of the
NSE stock market. Kumar et al. (2023a) have analized conditional variance objectively or empirically
estimates the price volatility spillover transmission in the daily returns of IPC Mexico index from
Mexico stock market using the GJR- GARCH model.
Leite & Lima (2023) revealed the extreme volatility of the spot price in Brazil. Institutional issues and
the rising proportion of renewable energy in the electrical mix are linked to this high volatility. Birau
et al. (2023) found that the GARCH (1, 1) model's perfect fit, which takes into account the impacts of
GARCH and ARCH, shows that the volatility in the Sweden market has persisted throughout time.
(Bonga, 2019) suggested that the volatility of the Zimbabwean stock market is modeled using GARCH
family. It was found that the EGARCH (1,1) model is the best suitable model. Spulbar et al. (2022)
have also observed volatility during the COVID-19 pandemic has been demonstrated to form a “V”
shape pattern where an unpredictable, sharp negative slope is generated. This was entirely different
from the pattern created during the global financial crisis. Bonga (2019) concluded that both positive
and negative shocks affect stock market returns differently. Both positive and negative news will boost
the volatility of stock market returns, but to varying degrees.
Research Gap
75
ANNALS OF THE “CONSTANTIN BRÂNCUȘI” UNIVERSITY OF TÂRGU JIU
LETTER AND SOCIAL SCIENCE SERIES
ISSN-P: 1844-6051 ~ ISSN-E: 2344-3677
https://alss.utgjiu.ro
2/2023
A significant research gap in the field of financial econometrics is filled by this work. While an
extensive amount of research has been done on forecasting stock market volatility, less focus has been
placed on modeling and comparative analysis of volatility in Switzerland. Furthermore, there is still a
dearth of research on the use of complicated GARCH models, including TGARCH, EGARCH, and
PARCH models, in this particular setting. By bridging this gap, our study provides investors,
policymakers, and financial analysts substantial understanding into the different dynamics and
asymmetric volatility patterns within these two different markets.
Research Methodology
In order to capture changes, volatility clusters, the fitness of econometric models, and volatility
patterns, the current work focuses on modeling the behavior of the Switzerland stock market using
data from4th January, 2000 to 9th November, 2023. We use GARCH models (Bollerslev, 1986). The
study used a sample number of 5994 daily observations for Swiss Stock Index representing the
Switzerland stock market. The time series data have been used for modeling volatility. The daily
returns were calculated using the log of the first difference of the daily closing prices. Before doing
any of these tests, the daily returns were compiled since volatility has been evaluated on return (rt).
The log of first difference of the daily closing price is used to compute the return series, which is as
follows:
𝑷𝒕
𝒓𝒕 = 𝒍𝒐𝒈
𝑷𝒕−𝟏
Where,
rt = Logarithmic daily return for time t,
Pt = Closing price at time t,
Pt−1 = Corresponding price in the period at time t − 1.
Before employing the GARCH models, we used some statistical techniques to evaluate stationarity.
The data's stationarity was officially evaluated using the Phillips-Perron and Augmented Dickey Fuller
tests statistic. The ARCH Lagrange Multiplier (LM) test was employed to investigate the presence of
heteroscedasticity in the residual series of the return data. Identifying heteroscedasticity is crucial in
choosing the appropriate GARCH model. To estimate the GARCH models (TGARCH, EGARCH,
PARCH, and Component ARCH (1,1). we utilized the E-Views 12 Econometrics package. This
software package provides robust tools for econometric modeling and time series analysis. The
selection of the most suitable GARCH model was based on the evaluation of four GARCH family
models: GARCH/TARCH, EGARCH, Component ARCH(1,1), and PARCH, all using the Student t's
Distribution.
Empirical Results and Discussion
In this paper, the daily closing prices of the Swiss stock index (SMI), over the period from 4th January,
2000 to 9th November, 2023 resulted in total observations of 5994 excluding public holidays. Various
descriptive statistics are calculated and exhibited in Table 1.1 providing 7933.704 mean with 1938.689,
degree of Standard Deviation. A high value of kurtosis2.402770 which is less than 3 indicates a
platykurtic distribution that is an apparent departure from normality while the skewness represents
positive value it indicating data has long right skewed distribution.
The Jarque-Bera statistic is a crucial normality test, the p-value of Jarque Bera is less than its critical
value of 5% signifying the data is non-normal.
Graph 1.1: Descriptive Statistics of the Swiss stock index
76
ANNALS OF THE “CONSTANTIN BRÂNCUȘI” UNIVERSITY OF TÂRGU JIU
LETTER AND SOCIAL SCIENCE SERIES
ISSN-P: 1844-6051 ~ ISSN-E: 2344-3677
https://alss.utgjiu.ro
2/2023
400
Series: SWISS_STOCK_INDEX__SMI__
Sample 1/04/2000 11/09/2023
Observations 5994
350
300
250
200
150
100
50
0
Mean
Median
Maximum
Minimum
Std. Dev.
Skewness
Kurtosis
7933.704
7939.710
12970.53
3675.400
1938.689
0.322911
2.402770
Jarque-Bera
193.2489
0.000000
4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 Probability
Source: Authors’ Calculation using Eviews 12
Graph 1.2: Movement Pattern of of the Swiss stock Index
Swiss stock index (SMI) closing price
14,000
12,000
10,000
8,000
6,000
4,000
2,000
00
02
04
06
08
Source: Authors’ Calculation using Eviews12
10
12
14
16
18
Graph 1.3: log returns of the Swiss stock Index
77
20
22
ANNALS OF THE “CONSTANTIN BRÂNCUȘI” UNIVERSITY OF TÂRGU JIU
LETTER AND SOCIAL SCIENCE SERIES
ISSN-P: 1844-6051 ~ ISSN-E: 2344-3677
https://alss.utgjiu.ro
2/2023
Swiss stock index (SMI) log returns
.12
.08
.04
.00
-.04
-.08
-.12
00
02
04
06
08
10
Source: Authors’ Calculation using Eviews 12
12
14
16
18
20
22
Graph 1.2 shows the movement patterns of the Swiss stockIndex’s Stationary Series during the
hypothetical period from 4th January, 2000 to 9th November, 2023. Graph 1.3 shows the graphical
presentation of the log returns of the presence of volatility clustering using the Swiss stock Index. In
order to estimate the volatility ofSwitzerlandstock market, checking the stationary is the first step in
the analysis of the return series (Maqsood et al., 2017). For this purpose, Augmented Dickey-Fuller
(Dickey & Fuller, 1979) test, Phillips Perron test (Phillips & Perron, 1988) and Kwiatkowski-PhillipsSchmidt-Shin test are used to establish the stationarity of the Swiss stock index sample data series.
The test results are presented with the help of following tables:
Table: 1.1: Unit root Test (Augmented Dickey-Fuller test, Phillips-Perron test and of Swiss
stock index
Null Hypothesis: D(SWISS_STOCK_INDEX__SMI__CLOSING_PRICE) has a unit root
Exogenous: Constant
Lag Length: 5 (Automatic - based on AIC, maxlag=33)
Augmented Dickey-Fuller test statistic
Test critical values:
1% level
5% level
10% level
t-Statistic
Prob.*
-33.92698
-3.431263
-2.861828
-2.566966
0.0000
Null Hypothesis: D(SWISS_STOCK_INDEX__SMI__CLOSING_PRICE) has a unit root
Exogenous: Constant
78
ANNALS OF THE “CONSTANTIN BRÂNCUȘI” UNIVERSITY OF TÂRGU JIU
LETTER AND SOCIAL SCIENCE SERIES
ISSN-P: 1844-6051 ~ ISSN-E: 2344-3677
https://alss.utgjiu.ro
2/2023
Bandwidth: 17 (Newey-West automatic) using Bartlett kernel
Phillips-Perron test statistic
Test critical values:
1% level
5% level
10% level
Adj. t-Stat
Prob.*
-76.14059
-3.431262
-2.861828
-2.566965
0.0001
Source: Authors’ Calculation using Eviews 12
Table 1.1 shows the Unit root Test (Augmented Dickey-Fuller test and Phillips-Perron test) of Swiss
stock index. Table 1.2 shows the p values of Augmented Dickey-Fuller test and Phillips-Perron test
statistic are less than 0.05which leads to reject the null hypothesis hence, the sample data were found
to be stationary since the probability values are significant at 10%, 5%, and 1% levels.
Testing for ARCH Lagrange Multiplier Effect:
It is crucial to look at the residuals for signs of heteroscedasticity. If conditional heteroskedasticity is
present, the results might be deceiving if it is not taken into consideration (Sokpo et al., 2017). The
ARCH Lagrange Multiplier (LM) test is employed to determine whether heteroscedasticity exists in
the return series' residual. Testing for conditional heteroskedasticity is crucial since if it's omitted
adopting GARCH-type models would be improper.
Table 1.3: Heteroskedasticity Test: ARCH
Heteroskedasticity Test: ARCH
F-statistic
Obs*R-squared
0.327156
0.327247
Prob. F(1,1662)
Prob. Chi-Square(1)
0.0000
0.0000
Source: Authors’ Calculation using Eviews 12
Table 1.3 shows the result of the ARCH-LM test for Swiss stock index. It inferred that data is highly
significant. The probability of F-statistic (0.0000) shows that p value is less than 0.05; the null
hypothesis (i.e., no ARCH effect) is rejected at 1% level. The results support to estimate GARCH
family models since, indicating the existence of ARCH effects in the residuals of time series models.
This indicates the series under consideration is variable, requiring volatility modeling to account for
volatility in the model.
Table 1.4: Selecting an appropriate model
Swiss stock Index
Estimated model
Akaike info criterion
Schwartz criterion
Log Likelihood
GARCH/TARCH
-6.541506
-6.534799
19604.35
EGARCH
-6.584794
-6.576969
19735.04
PARCH
-6.588085
-6.579142
19745.90
79
ANNALS OF THE “CONSTANTIN BRÂNCUȘI” UNIVERSITY OF TÂRGU JIU
LETTER AND SOCIAL SCIENCE SERIES
ISSN-P: 1844-6051 ~ ISSN-E: 2344-3677
https://alss.utgjiu.ro
2/2023
Component ARCH
(1,1)
-6.541095
Source: Authors’ Calculation using Eviews 12
-6.532152
19605.12
Table 1.4 Depicts four models of GARCH family. PGARCH with Student t's Distribution has the
lowest Akaike info criterion with -6.588085and Schwartz criterion with -6.579142apart from that
maximum Log Likelihood with 19745.90when compared to the other three. As a result, this model is
thought to be the best one. The results of the selected PARCH Model for the Swiss stock Index are
shown in the table below.
Table 1.5: PARCH with Student's t distribution Error Construct of Swiss stock index
Dependent Variable: SWISS_STOCK_INDEX__SMI__LOG_RETURNS
Method: ML ARCH - Student's t distribution (BFGS / Marquardt steps)
Date: 11/23/23 Time: 21:19
Sample (adjusted): 1/06/2000 11/09/2023
Included observations: 5992 after adjustments
Convergence achieved after 174 iterations
Coefficient covariance computed using outer product of gradients
Presample variance: backcast (parameter = 0.7)
@SQRT(GARCH)^C(7) = C(3) + C(4)*(ABS(RESID(-1)) - C(5)*RESID(
-1))^C(7) + C(6)*@SQRT(GARCH(-1))^C(7)
Variable
C
SWISS_STOCK_INDEX__SMI__L
OG_RETURNS(-1)
Coefficient
Std. Error
z-Statistic
Prob.
0.000163
0.000101
1.613265
0.1067
0.010790
0.013234
0.815317
0.4149
Variance Equation
C(3)
C(4)
C(5)
C(6)
C(7)
0.000325
0.087259
0.999961
0.902459
0.977956
0.000132
0.005035
8.14E-07
0.006362
0.082126
2.463955
17.32881
1228087.
141.8461
11.90803
0.0137
0.0000
0.0000
0.0000
0.0000
T-DIST. DOF
9.795678
0.910934
10.75344
0.0000
R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
0.000304
0.000137
0.011285
0.762885
19745.90
1.973995
Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
6.57E-05
0.011286
-6.588085
-6.579142
-6.584979
Source: Authors’ Calculation using Eviews 12
Above table are representing the PARCH model with Student's t distribution error construct of
Swissstock Index. Since Probabilities are lower than 0.05, the constant (C) is considered significant.
80
ANNALS OF THE “CONSTANTIN BRÂNCUȘI” UNIVERSITY OF TÂRGU JIU
LETTER AND SOCIAL SCIENCE SERIES
ISSN-P: 1844-6051 ~ ISSN-E: 2344-3677
https://alss.utgjiu.ro
2/2023
Graph 1.4: Estimating volatility patterns using PARCH models of Swiss Stock Index
.10
Forecast: SWISS_STOCF
Actual: SWISS_STOCK_INDEX__SMI__LO...
.05
Forecast sample: 1/04/2000 11/09/2023
.00
Adjusted sample: 1/06/2000 11/09/2023
Included observations: 5992
-.05
-.10
Root Mean Squared Error
0.011283
Mean Absolute Error
0.007746
Mean Abs. Percent Error
NA
Theil Inequality Coef. 0.982095
-.15
00 02 04 06 08 10 12 14 16 18 20 22
SWISS_STOCF
Bias Proportion
0.000075
Variance Proportion
0.978832
Covariance Proportion
± 2 S.E.
.0025
0.021093
Theil U2 Coefficient
NA
Symmetric MAPE
187.1297
.0020
.0015
.0010
.0005
.0000
00 02 04 06 08 10 12 14 16 18 20 22
Forecast of Variance
Source: Authors’ Calculation using Eviews 12
We can forecast the volatility of the Switzerland Stock Exchange composite indices using the
aforementioned methodology using a data set of 5994 days. The Graph 1.4 demonstrates the
anticipated uneven price changes of the Switzerland Stock Exchange.
Conclusions
To forecast variance using financial time series data, we used model i.e., Generalized Autoregressive
Conditional Heteroskedasticity (GARCH), which is specifically designed for volatility forecasting.
The main aim of the present study is to modeling the behavior of the Switzerland stock market using
data from 4th January, 2000 to 9th November, 2023. Through the application of GARCH family models
which, include GARCH/TARCH, EGARCH, Component ARCH (1,1), and PARCH. The analysis
centered on log returns derived from the Swiss Stock Index. We used some statistical techniques to
evaluate stationarity. The Augmented Dickey Fuller test, and Phillips-Perron test, and Kwiatkowski81
ANNALS OF THE “CONSTANTIN BRÂNCUȘI” UNIVERSITY OF TÂRGU JIU
LETTER AND SOCIAL SCIENCE SERIES
ISSN-P: 1844-6051 ~ ISSN-E: 2344-3677
https://alss.utgjiu.ro
2/2023
Phillips-Schmidt-Shin test statistic. Findings indicating that the sample data were stationary. The
ARCH Lagrange Multiplier (LM) test was employed to investigate the presence of heteroscedasticity
in the residual series of the return data.
The results of this test revealed the presence of ARCH effects in the residuals of our time series models.
The selection of the most suitable GARCH model was based on the evaluation of four GARCH family
models that is GARCH/TARCH, EGARCH, Component ARCH (1,1), and PGARCH using Student
t's distribution. As a result, PARCH Model were selected with the help of the lowest Akaike info
criterion, Schwartz criterion and maximum Log Likelihood.
The results obtained through the application of the PARCH model, as showcased in Table 1.5, provide
valuable insights into the dynamics of theSwiss Stock Index. The conclusions and methods discussed
in this study provide a strong basis for further research, enhancing our capacity to predict market
movements and make wise choices in a volatile financial environment. This study highlights the
significance of accurately and meticulously simulating stock market behavior in addition to adding to
the corpus of knowledge in financial econometrics. By using sophisticated GARCH models and doing
thorough statistical analyses, we have been able to gain important insights into the workings of the
Switzerland stock markets, which have helped us to better, comprehend the complex field of stock
market volatility prediction.
82
ANNALS OF THE “CONSTANTIN BRÂNCUȘI” UNIVERSITY OF TÂRGU JIU
LETTER AND SOCIAL SCIENCE SERIES
ISSN-P: 1844-6051 ~ ISSN-E: 2344-3677
https://alss.utgjiu.ro
2/2023
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
Badarla, S., Nathwani, B., Trivedi, J., Spulbar, C., Birau, R., Hawaldar, I.T., Minea, E.L. (2021) Estimating
fluctuating volatility time series returns for a cluster of international stock markets: A case study for Switzerland,
Austria, China and Hong Kong, Physics AUC (Annals of the University of Craiova, Physics), vol. 31, 43-52.
Baillie, R. T., Bollerslev, T., & Mikkelsen, H. O. (1996). Fractionally integrated generalized autoregressive
conditional heteroskedasticity. Journal of Econometrics , 74, 3–30.
Birau, R., Spulbar, C., Trivedi, J., & Florescu, I. (2021) Modeling volatility in the stock markets of Spain and
Hong Kong using GARCH family models in the context of COVID - 19 pandemic, Revista de Științe Politice.
Revue des Sciences Politiques, 72, 3 – 21.
Birau, R., Trivedi, J., Baid, R., Florescu, I., & Simion, M. L. (2023). Estimating volatility patterns using GARCH
models: A case study on Swedish stock market. Revista de Științe Politice. Revue des Sciences Politiques , 78,
50-59.
Bollerslev, T. (1986). Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics , 31
(3), 307–327.
Bonga, W. G. (2019). Stock Market Volatility Analysis using GARCH Family Models: Evidence from Zimbabwe
Stock Exchange. Munich Personal RePEc Archive , 2-13.
Choudhry, T., & Wu, H. (2008). Forecasting ability of GARCH vs kalman filter method: evidence from daily UK
time-varying beta. Journal of Forecasting , 27, 670–689.
Dickey, D. A., & Fuller, W. A. (1979). Distribution of the Estimators for Autoregressive Time Series with a Unit
Root. Journal of American Statistical Association , 74 (366), 427–31.
Engle, R. F. (1982). Autoregressive conditional heteroscedasticity with estimates of the variance of United
Kingdom inflation. Econometrica , 50 (4), 987–1007.
Kim, H. Y., & Won, C. H. (2018). Forecasting the Volatility of Stock Price Index: A Hybrid Model Integrating
LSTM with Multiple GARCH-Type Models. Expert Systems With Applications , 1-33.
Kumar, S., Anand, A., Birau, R., Meher, B. K., Kumar, S., & Ion, F. (2023a). Temporal Analysis of Mexico Stock
Market Index Volatility using GJR-GARCH model. Revista de Științe Politice. Revue des Sciences Politiques ,
79, 46-56.
Kumar, S., Meher, B.K., Birau, R., Simion, M.L., Abhishek, A., & Manohar, S. (2023b) Quantifying Long‐Term
Volatility for Developed Stock Markets: An Empirical Case Study Using PGARCH Model on Toronto Stock
Exchange (TSX), Annals of “Dunarea de Jos” University of Galati Fascicle I. Economics and Applied Informatics
Years XXIX – no2/2023, DOI https://doi.org/10.35219/eai15840409338, ISSN-L 1584-0409, ISSN-Online 2344441X, 61-68.
Kumar, S., Meher, B. K., Birau, R., Simion, M. L., Anand, A., & Kumar, S. (2023c). Long-term volatility
forecasting of Brazilian stock index behavior using suitable GARCH family models. Revista de Științe Politice.
Revue des Sciences Politiques , 79, 9-24.
Leite, A. L., & Lima, M. V. (2023). A GARCH Model to Understand the Volatility of the Electricity Spot Price
in Brazil. International Journal of Energy Economics and Policy , 13 (5), 332-338.
Maqsood, A., Safdar, S., Shafi, R., & Lelit, N. J. (2017). Modeling Stock Market Volatility Using GARCH
Models: A Case Study of Nairobi Securities Exchange (NSE). Open Journal of Statistics , 7, 369-381.
Meher, B. K., Hawaldar, I. T., Mohapatra, L., Spulbar, C., & Birau, R. (2020). The Effects of Environment,
Society and Governance Scores on Investme Investment Returns and Stock Market Volatility. International
Journal of Energy Economics and policy , 10 (4), 234-239.
Phillips, P. C., & Perron, P. (1988) Testing for a Unit Root in Time Series Regression. ’ Biometrika , 75 (2), 335346, Published By: Oxford University Press.
Sokpo, J. T., Iorember, P. T., & Usar, T. (2017). Inflation and Stock Market Returns Volatility: Evidence from
the Nigerian Stock Exchange 1995Q1-2016Q4: An E-GARCH Approach. Munich Personal RePEc Archive , 119.
Spulbar, C., Birau, R., & Imran, Z. A. (2020). Analyzing Short Term Momentum Effect on Stock Market of Hong
Kong. An Empirical Case Study,. “Ovidius” University Annals, Economic Sciences Series , XIX (2/2019), 889894.
Spulbar, C., Birau, R., Trivedi, J., Hawaldar, I. T., & Minea, E. L. (2022). Testing volatility spillovers using
GARCH models in the Japanese stock market during COVID-19, Investment Management and Financial
Innovations , 19 (1), 262-273.
Spulbar, C., Birau, R., Trivedi, J., Simion, M.L., Baid, R. (2023) Assessing Volatility Patterns using GARCH
Family Models: A Comparative Analysis Between the Developed Stock Markets in Italy and Poland, Annals of
83
ANNALS OF THE “CONSTANTIN BRÂNCUȘI” UNIVERSITY OF TÂRGU JIU
LETTER AND SOCIAL SCIENCE SERIES
ISSN-P: 1844-6051 ~ ISSN-E: 2344-3677
https://alss.utgjiu.ro
2/2023
“Dunarea de Jos” University of Galati Fascicle I. Economics and Applied Informatics Years XXIX – no1/2023,
pp.5-11, ISSN-L 1584-0409, ISSN-Online 2344-441X, DOI https://doi.org/10.35219/eai15840409314.
22. Sun, H., & Yu, B. (2019). Forecasting Financial Returns Volatility: A GARCH-SVR Model. Computational
Economics .
23. Vedat, A. (1989). Conditional heteroskedasticity in time series models of stock returns: Evidence and forecasts.
Journal of Business , 62, 55–80.
84