INTERNATIONAL JOURNAL OF RESEARCH IN BUSINESS AND SOCIAL SCIENCE 12(1)(2023) 204-211 Research in Business & Social Science IJRBS VOL 12 NO 1 (2023) ISSN: 2147-4478 Available online at www.ssbfnet.com Journal homepage: https://www.ssbfnet.com/ojs/index.php/ijrbs Forecasting volatility in oil returns using asymmetric GARCH models: evidence from Tanzania Laban Gasper Letema (a)* (a) (b) Haika Andrew Mbwambo (b) Assistant Lecturer, College of Business Education, Dodoma, Tanzania Assistant Lecturer, College of Business Education, Mwanza, Tanzania ARTICLEINFO ABSTRACT Article history: Crude oil is, without a doubt, one of the most significant commodities in the modern world. The highly contagious coronavirus, the conflict between Ukraine and Russia, and not to mention the unusual turn of events worldwide have all significantly impacted crude oil prices. Since oil is required for all critical economic activities, such as production and transportation, a forecast for crude oil prices is essential. Using a range of GARCH models at such an intense time, this study attempted to close this gap by forecasting crude oil volatility. To forecast the returns of Brent crude oil prices from January 2002 to February 2022, this study uses a family of GARCH models. In the respective family of models, GJRGARCH (1,1) was the most effective in predicting the volatility of crude oil prices. The GJRGARCH model was chosen since it had a higher likelihood value and a lower information criteria value. A diagnostic check was done to evaluate the produced model further to ensure that the proposed model was good enough for forecasting crude oil volatility. The study suggests employing the GJRGARCH technique to predict future fluctuations in exceptional circumstances.. Received 29 October 2022 Received in rev. form 22 Dec. 2022 Accepted 24 January 2023 Keywords: Brent, crude oil, GJRGARCH, forecasting, returns, volatility JEL Classification: F65 © 2023 by the authors. Licensee SSBFNET, Istanbul, Turkey. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). Introduction Undoubtedly, one of the most important commodities in today’s globe is crude oil. Heating oil, diesel, motor gasoline, and jet fuel for automobiles, ships, and aircraft are the most common types of oil that come to mind when we think of it (Selvi et al., 2018). Many of the things we use every day depend on oil for their manufacturing and transportation. A significant influence on the other industries comes from the oil producing sector. Any change in the price of petroleum products has an enormous impact on the costs of other items produced and even the health of the economy (Fondo, & Otulo, 2021).Oil price swings have a cascading impact on our daily lives, affecting things like food supply, detergents, prescription drugs, and household appliances, to name a few. Similar to any other commodity, its price changes in reaction to market forces of demand and supply. Changes in the oil prices will either negatively affect many economic sectors or positively impact others as they rise or fall (Mensah, 2015). Making crude oil one of the main economic issues and a major topic of debate in global economic policy. Given the last two decades, it is clear that the financial crisis caused by the housing bubble had a negative impact on oil prices, which went from US$ 133.88 per barrel in June 2008 to under $40 per barrel within a few months of the crisis. Following that, the prices rose till they reached $100 in 2014, but this did not persist for long, and they dropped below $30 in 2016 as a result of the increased supply of crude oil. Due to the impacts of Covid 19, prices in 2020 decreased to their lowest point in 20 years, reaching $ 16.55 in April. But as a result of the situation in Ukraine, prices increased once more. These shifts in the price of crude oil internationally undoubtedly have an impact on nations that rely significantly on imported crude oil (Nyongesa and Wagala, 2016; Rodhan and Jaaz 2022). Recently Covid 19 had a negative impact on oil prices, the prices of crude oil in 2020 decreased to their lowest point in 20 years, reaching $ 16.55 in April. But as a result of Ukraine war, prices increased once more. These shifts in the price of crude oil internationally undoubtedly have an impact on nations that rely significantly on imported crude oil (Rodhan & Jaaz, 2022). * Corresponding author. ORCID ID: 0000-0002-4284-1911 © 2023 by the authors. Hosting by SSBFNET. Peer review under responsibility of Center for Strategic Studies in Business and Finance. https://doi.org/10.20525/ijrbs.v12i1.2308 Letema & Mbwambo, International Journal of Research in Business & Social Science 12(1) (2023), 204-211 Tanzania has been exporting natural gas for more than 50 years and is a significant producer of the fuel. Songo Songo Island (Lindi Region) made the first natural gas finding in Tanzania, which was thereafter followed by Mnazi Bay (Mtwara Region). However, Tanzania neither produces crude oil nor have any recent commercial oil discovery (Saxena & Ndule, 2020). An average of 35,000 barrels of refined oil products are consumed daily in Tanzania, all of which are imported. The consumption of crude oil and petroleum products makes up the greatest portion of all commercial energy consumption, and it has been expanding quickly since 2010 along with an increase in motorization (JICA, 2022). However, their supply is fully dependent on imports. The Bank of Tanzania reports that in addition to an increase in oil imports, domestic prices also climbed in line with prices on the international market (BANK OF TANZANIA, 2022) taking into account the effects of COVID-19 and the aftermath of the war in Ukraine. Tanzania relies heavily on crude oil imports for the majority of its industrial and socioeconomic operation (Shakiru & Liu, 2022). Oil is used to transport goods and services; its price has an impact on the entire economy. The price of goods and services tends to rise when the price of oil rises. Moreover, when gasoline and diesel prices rise, it affects everyone which causes inflation (Saxena & Ndule, 2020). The rapid growth of Tanzania’s economy has necessitated an increase in the demand for crude oil which has the multiplier effect on the economy. Several authors examined the asymmetric relationship between oil price shocks and macroeconomic fluctuations in Tanzania and found that crude oil prices have an impact on GDP and inflation (Shakiru & Liu, 2022), on explaining the variability of Tanzanian shillings’ exchange rates and even on stock market performance (Kasongwa & Minja, 2022).It is thus crucial to conduct a thorough analysis and forecast of crude oil prices since they have a multiplier effect on the production, distribution of goods and the economy at large. A stochastic modeling approach that incorporates the time-dependent structure present in the time series crude oil price data can be used to study the dynamics and evolution of crude oil prices. One of the key components of an economic analysis is forecasting. Past observations can be utilized to forecast future values by identifying suitable models to reflect the data. There are several analytical techniques that have gained a lot of attraction recently in forecasting crude oil. These are such as GARCH-type models as applied by (Ahmed & Shabri, 2014; Wacuka Ng’ang’a & Oleche, 2022; Deebom & Essi, 2017), Support Vector Machine (SVM) that is used to forecast data of high volatility (Mensah, 2015) and Autoregressive Integrated Moving Average (ARIMA) popularly known as Box 1 Jenkins Methodology (Shambulingappa et al., 2020; Rodhan & Jaaz, 2022). GARCH family models are generally considered to be the most efficient methods since they are more capable of forecasting than other forecasting models (Shabani et al. 2016; Haque & Shaik, 2021). They are considered essential for determining the volatility of a number of commodities ( Charles, & Darné, 2021). The GJR-GARCH model also distinguishes itself from other forecasting models by accounting for the effects of leverage, volatility clustering, and leptokurtosis in the time series analysis (Khan, Khan & Sheeraz, 2019). Additionally, both symmetric and asymmetric models are found to be effective at capturing volatility (Ekong, & Onye, 2017). Nevertheless, there have been numerous attempts to forecast the price of crude oil. There are, however, few studies that have attempted to address the issue of forecasting crude oil prices in this period of exceptional circumstances in the Tanzanian context, where macroeconomic indicators including crude oil prices have been affected by recent shocks. The spread of corona virus caused a dramatic drop in oil prices, while the current Ukraine-Russian war has caused a sharp rise in crude oil prices. To fill this gap, the current study goes on to predict crude oil volatility using GARCH in such an extreme period. This study’s primary goal is to support the growth of the energy markets through an analysis of the models that are used to predict and simulate the volatility of crude oil prices. Prices of other commodities are directly impacted by changes in oil prices. It is crucial to quantify the risk of investment loss as investors try to predict future prices so they can make investment decisions. Market players will be able to make informed decisions and develop trading policies with the aid of a proper study of the patterns in crude oil prices. As a result, traders who might want to buy the commodity as well as investors looking for a return will have access to the energy market. By simulating the volatility of crude oil prices, this study hopes to contribute to this. The emphasis is on price volatility models, where various GARCH models are examined to determine which model is best. The rest of the paper is structured as follows: the literature review, which primarily examines the research that other scholars have done and their findings, comes next. Followed by a thorough, step-by-step description of the model and the explanation of the results. The final section concludes up the paper and include recommendations for further research. Literature Review Wacuka Ng’ang’a and Oleche (2022) employed in this study for forecasting Brent crude oil price volatility a family of GARCH models. The IGARCH T-distribution model was found to be the best model for forecasting crude oil prices. Similar conclusions were reached by Cerović Smolović, et.al (2017), who investigated the performance of eight GARCH models and discovered that the TS GARCH, T GARCH, and EGARCH models perform the best at predicting volatility in the Montenegrin emerging market. Suleiman, Alabi, Issa, Usman, and Adamu (2015) looked at the most effective GARCH and ARIMA models for accurately forecasting the price of crude oil in Nigeria. The 189 monthly crude oil price observations used in this analysis covered the period from January 1998 to September 2013. Based on factors like AIC, HQC, and SIC, their study evaluated fifteen (15) models and chose the top ARIMA and GARCH models. The model with the least values of the criteria was deemed to be the best model. According to their findings, the best model for predicting the crude oil price data series was GARCH (2, 1). Their projection, which was developed 205 Letema & Mbwambo, International Journal of Research in Business & Social Science 12(1) (2023), 204-211 for a period of six months, indicates a sharp increase in the price of crude oil when compared to historical averages. Nonetheless, Herrera, Hu, & Pastor, (2018) methodology follows the GARCH and GARCH-X models. The results show that there exists a positive effect between the deviations and food price volatility. On the contrary, for other authors that did a hybrid of ARIMA models with other models like (Mensah, 2015) show that ARIMA model is not able to capture the volatility inherent in the crude oil price for an accurate forecast. (Fondo et al., 2021) also made predictions of petroleum prices. Prediction was made for the next twelve months. Due to data volatility both ARIMA and the VAR model were applied. The VAR model had the least error. Hence according to their study VAR is a better model for predicting petroleum prices in Kenya. Ahmed and Shabri (2014) also forecasted crude oil price based of three techniques, Support Vector Machines (SVM) in comparison to the performance to ARIMA and GARCH. In their study data on crude oil price of West Texas Intermediate (WTI) was used. The results revealed that SVM method outperforms the other two in terms of forecast accuracy as it achieved the smallest forecast error judging by their RMSE and MAE followed by ARIMA then GARCH. The results reviewed that the proposed SVM method outperforms the others. (Shah, 2020) also had similar conclusions on SVM method. Moreover, (Liu, Li, Qianjie Geng & Yudong Wang, 2022) analyzed the oil volatility forecasting focusing from the viewpoint of investors who are more concerned with economic predictability than statistical predictability. They employed the most popular GARCH-type models, however no single volatility model consistently outperformed its competitors. They further recommended that it would be feasible to find a small group of models that, when combined, provide more accurate findings at a particular degree of confidence. Research and Methodology Research design A time series design was used in this study. This design was chosen due to the fact that crude oil prices tend to fluctuate over time. Using monthly crude oil prices from January 2002 to April 2022 as a suitable time series, the family of GARCH models were fitted to produce a stable model for crude oil prices using historical data. Data description and source Monthly Brent Crude oil spot prices spanning from January 2002 to April 2022 from central bank of Tanzania website were used in this study for modeling and forecasting. This information was chosen because Brent crude oil is considered as the benchmark for crude oils in Europe and Africa. Crude oil is also traded by its own or its price is reflected by the price of other types of crude oil. The availability of data in the given time frame was another consideration in this selection. Data analysis R Statistical Software was used to analyze the obtained time series data in order to produce projections that have been used to predict the future volatility returns of crude oil price. Analysis also involved the creation of diagnostic test values and time series plots which were necessary to justify the suitability of the model for the study. Stationarity of the data Many time series techniques make the assumption that the data is stationary. A stationary process is one in which the series tend to return to back to mean (Haque & Shaik, 2021). This denotes a trendless, flat series with no variance variations over time. Differencing is used to make a process stationary if it is discovered to be non- stationary (Wacuka Ng’ang’a & Oleche, 2022). ARCH model The generalized form of the ARCH model of order (p) is given by rt = µ + ϵt, t = 1, 2, 3, . . . , T Where; T = number of observations, ϵt = residuals and µ = mean of the time series (rt). Nevertheless, the residuals in the ARCH process shows that the following assumptions are made about the residuals: 𝜀𝑡 = δ𝑡 z𝑡 The series of σ2 is modelled by the following equation zt ∼ N(0, 1) 2 𝛿𝑡 2 = 𝜔𝑂 + ∑𝑝𝑖=1 𝛼𝑖 𝜀𝑡−𝑖 + 𝜀𝑡 In equation 3, ωo > 0, αi 0 and ϵt is assumed to follow a standard normal or student t-distribution ≥ 206 Letema & Mbwambo, International Journal of Research in Business & Social Science 12(1) (2023), 204-211 GARCH model To capture volatility in the series the symmetric GARCH model was employed in this study. The volatility in a return series is more accurately represented by the symmetric GARCH model than the ARCH model (Saltik, Degirmen, & Ural, 2016). The GARCH (p,q) model is expressed and defined as follows: 2 2 𝛿𝑡 2 = 𝜔𝑂 + ∑𝑝𝑖=1 𝛼𝑖 𝜀𝑡−𝑖 + ∑𝑝𝑖=1 𝛽𝑗 𝛿𝑡−𝑗 + 𝜀𝑡 The model assumes that shocks are distributed as independent and identical. The constraints of the model are; ωo > 0, αi ≥ 0, βi > 0 and αi + βi < 1 GJRGARCH Model Glosten et al. (1993) established the nonlinear model GJRGARCH to identify the asymmetric leverage effect, which allows us to distinguish between positive and negative news effects over series. Glosten et al. (1993) added a dummy variable (Di) to the Bollerslev GARCH model in order to analyze these effects (1986) IGARCH Model 2 2 𝛿𝑡 2 = 𝜔𝑂 + ∑𝑝𝑖=1(𝛼 𝑖 + 𝐷𝑖 𝛾𝑖 )𝜀𝑡−𝑖 + ∑𝑝𝑖=1 𝛽𝑗 𝛿𝑡−𝑗 + 𝜀𝑡 Engle and Bollerslev’s (1986) developed an Integrated Generalized Conditional Heteroscedasticty (IGARCH) class of non linear model which is mathematically written as follows: 2 2 𝛿𝑡 2 = 𝜔𝑂 + ∑𝑝𝑖=1 𝛼𝑖 𝜀𝑡−𝑖 + ∑𝑝𝑖=1 𝛽𝑗 𝛿𝑡−𝑗 + 𝜀𝑡 The only differences between the IGARCH model and the standard GARCH model are the parameter restrictions, which are ωo > 0 and αI + γj = 1. Shocks and innovations indicate continuity in this aspect (Deebom & Essi, 2017). GARCH Model Exponential GARCH (EGARCH), created by Nelson (1991), is another nonlinear GARCH model that takes asymmetry and leverage effects into consideration. The ability of EGARCH to decompose positive and negative shocks/innovations from one another and take logarithms of conditional variance, hence preventing negative values of conditional variance, are the main differences between EGARCH models and Bollerslev’s (1986) GARCH model (Haque & Shaik, 2021). The model can be formulated as follows 𝑝 log (𝛿𝑡 2) = 𝜔𝑂 + ∑𝑖=1 𝛼𝑖 |𝜀𝑡−𝑖 | + 𝛾𝑖 𝜀𝑡−𝑖 𝛿𝑡−𝑖 2 𝑝 2 + ∑𝑖=1 𝛽𝑗 𝛿𝑡−𝑗 + 𝜀𝑡 The model does not place any restrictions on α, γ and β. This appears to be a bene- fit of the model. It is feasible to discuss leverage and asymmetry effects because the parameter γ differs from zero and accepts values between confidence intervals. Additionally, if γ > 0 positive shocks/innovations have a considerably greater impact on conditional variance than negative shocks (Deebom & Essi, 2017). Findings & Discussion Stationarity features of crude oil prices For the initial investigation, time series plots for the given series were employed as shown in figure 1 Figure 1: Monthly crude oil prices 207 Letema & Mbwambo, International Journal of Research in Business & Social Science 12(1) (2023), 204-211 Visual examination of the time plot reveals that the mean and variance are evidently non-constant, indicating that the data is not stationary. Returns This study examined the volatility of crude oil prices using returns from the same time frame. To analyse the volatility of crude oil prices, we use oil returns for the same period. Prices are constrained to be non-negative and typically have a unit root, whereas logreturns can have any value, making them simpler to model and allowing for stationarity. This is why returns are preferable for modeling volatility (Sekati, Tsoku, & Metsileng, 2020). As seen in figure 2, which shows the log returns, time plots are utilized to identify the observable aspects of the returns. Figure 2: Plot of monthly crude oil returns This graph demonstrates the volatility clustering in the crude oil price monthly re turns. The crude oil price exhibits a strong increase followed by a sudden decline. Figure 2 shows that low volatility periods are followed by low volatility periods, while high volatility times are followed by high volatility periods. This shows the existence of volatility clustering in the crude oil returns. Therefore, we can further test for existence of Arch effect in the return of oil price data Testing for the presence of ARCH effect The Lagrange multiplier (LM) test was used to determine whether the ARCH effect was present in the residuals (1982). The hypothesis of the test is as follows. H0 = α 1 = α 1 = α 2 · · · = α n = 0 H1 = α1 = α1 = α2 · · · = αn ̸= 0 The findings of the ARCH-LM tests offer convincing evidence against the null hypothesis, as shown by the p-value (2.386e-09) which is less than 0.05. As a result, an ARCH or GARCH model should be used to model the return of crude oil prices. Model fitting Although estimating the order (p, q) to use when modeling might be challenging, various research have demonstrated that the predictive power of models does not always rise with the increase in order. In this study EGARCH( 1, 1), IGARCH (1, 1), GARCH (1, 1), FIGARCH (1, 1) and GJR GARCH (1, 1) models were fitted to the data. Estimated parameters of the model under normal distribution Table 1: Estimated parameters - normal distribution µ ar1 ma1 ω α β γ 208 GARCH (1, 1) 0.016579 -0.868125 0.946304 0.004289 0.501119 0.000001 - GJRGARCH (1, 1) 0.010851 0.107265 0.109330 0.004238 0.121638 0.000000 0.786152 EGARCH (1, 1) 0.009278 0.116109 0.126572 -2.252710 -0.221493 0.549841 0.621875 IGARCH (1, 1) 0.012387 0.145124 0.077127 0.001965 0.741775 0.258225 - Letema & Mbwambo, International Journal of Research in Business & Social Science 12(1) (2023), 204-211 Log likelihood and information criteria under normal distribution Table 2: Log likelihood and information criteria - normal distribution LogLikelihood Akaike GARCH (1, 1) 258.9608 -2.0820 GJRGARCH (1, 1) 265.0996 -2.1243 EGARCH (1, 1) 262.2614 -2.10094 IGARCH (1, 1) 256.6468 -2.0712 Bayes -1.9957 -2.0237 -2.0003 -1.9993 Shibata -2.0832 -2.1259 -2.1025 -2.0720 Hannan-Quinn -2.0472 -2.0837 -2.0604 -2.0422 Estimated parameters of the model under student’s t distribution Table 3: Estimated parameters - student t distribution µ ar1 ma1 Ω Α Β Γ Shape GARCH (1, 1) 0.018610 0.058657 0.119633 0.004140 0.537648 0.000001 12.760964 GJRGARCH (1, 1) 0.011934 0.098147 0.115046 0.004221 0.133355 0.000000 0.749338 33.502677 EGARCH (1, 1) 0.010186 0.111853 0.119279 -2.055590 -0.187977 0.591024 0.615490 19.327020 IGARCH (1, 1) 0.014028 0.113023 0.091675 0.001863 0.703221 0.296779 9.625200 Log likelihood and information criteria under student t distribution Table 4: Log likelihood and information criteria - student t distribution LogLikelihood Akaike GARCH (1, 1) 260.982 -2.0904 GJRGARCH (1, 1) 265.2759 -2.1175 EGARCH (1, 1) 262.9592 -2.0984 IGARCH (1, 1) 258.7 -2.0798 Bayes -1.9898 -2.0025 -1.9834 -1.9936 Shibata -2.0920 -2.1196 -2.1005 -2.0810 Hannan-Quinn -2.0499 -2.0712 -2.0521 -2.0451 Model of the Best Fit The model chosen as the best is the one with the lower information criteria value. The GJRGARCH (1, 1) model has the highest log likelihood and the lowest information criteria values for all examined distributions, according table 2 and 4. The findings are consistent with earlier research of (Wacuka Ng’ang’a and Oleche, 2022) but contrary to the results reported by (Ahmed and Shabri 2014; Shah, 2020). Since the coefficient of this models that capture the leverage effect is positive, it is clear that negative shocks (bad news) increase market volatility. This demonstrates that of the studied models, the GJRGARCH (1, 1) following normal distribution has the greatest fit. After fitting the model, we run a diagnostic test to determine whether the chosen model is stable enough to predict volatility in the crude oil data. Model Diagnostics Heteroskedasticity Test There is a need for additional confirmatory testing to determine whether the chosen model is suitable for forecasting, and this is to test for the presence of the ARCH effect. 209 Letema & Mbwambo, International Journal of Research in Business & Social Science 12(1) (2023), 204-211 Table 5: ARCH-LM Results Lag 3 Lag 5 Lag 7 Statistic 0.02261 0.37022 1.41502 Shape 0.500 1.440 2.315 Scale 2.000 1.667 1.543 P-Value 0.8805 0.9211 0.8383 This is done to examine homoscedasticity in the residuals generated by the model. It was estimated by using the ARCH-LM model, and the findings, which are displayed in table 5, indicate that there is no ARCH effect in the residuals. Ljung-Box Test on Standardized Residuals The outcome of the Box-Ljung test demonstrates that there is no serial association between the residuals. As a result, it appears that the GJRGARCH (1, 1) model may accurately capture the volatility in the returns of crude oil prices. Discussion In this work, a family of GARCH models were used and their degree of accuracy was compared. In order to further evaluate the produced model, a diagnostic check was done to ensure that the proposed model was good enough for forecasting crude oil volatility. According to the study, the hybrid GJRGARCH (1,1) model performed better than the other models in the family of GARCH models. The performance of the suggested hybrid model was also evaluated using data on Brent crude oil prices, and it likewise shown higher efficacy when compared to other models employed in this study. The suggested model GJRGARCH (1,1) beat its rival models in terms of forecasting accuracy measures. Similar conclusions were reached by (Wacuka Ng’ang’a and Oleche, 2022; Herrera & Pastor, 2018), but contrary to the results presented by Liu, et.al (2022); Wei, Wang, & Huang (2010) who analyzed the oil volatility using the most popular GARCH-type models and found that no single volatility model consistently outperformed its competitors. Conclusions Using a range of GARCH models at such an intense time, this study aimed at forecasting crude oil volatility. In order to forecast the returns of Brent crude oil prices for the period from January 2002 to February 2022, this study used a family of GARCH models. The presented findings demonstrates that the non-symmetric GJRGARCH (1, 1) model is a suitable model to predict crude oil price volatility. Since the coefficient of this models that capture the leverage effect is positive, it is clear that negative shocks (bad news) increase market volatility. The analysis comes to the conclusion that the GARCH (1, 1) model is the most accurate model for predicting the price of crude oil. The study recommends that further studies especially for the Tanzanian context should focus in evaluating the cause of oil volatility from the perspective of portfolio management and option pricing which can further aid investment decisions. Oil plays a crucial role in contemporary business, so it is important to consider the consequences for asset pricing and the predictive content of oil volatility. Furthermore, the researcher lacked the tools required to include a variety of volatility models in this study. There was a gap for other types of models, like multifractal models, because the R-packages utilized in the study could only support GARCH models. It is advised to conduct more research in order to include models like Markov Switching Multifractal Models in the category of volatility models employed. For this, R packages that can include these models would need to be created. For this, it would be necessary to develop R packages that can accommodate these models. Acknowledgement Author Contributions: Conceptualization, L.G.L and H.A.M; methodology, L.G.L; validation, L.G.L; formal analysis, L.G.L and H.A.M; investigation, L.G.L resources, L.G.L and H.A.M; writing—original draft preparation, L.G.L and H.A.M; writing—review and editing, L.G.L and H.A.M. Informed Consent Statement: Ethical review and approval were waived for this study, due to that the research does not deal with vulnerable groups or sensitive issues. Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available due to restrictions. Conflicts of Interest: The authors declare no conflict of interest. References Ahmed, R. A., & Shabri, A. B. (2014). Daily crude oil price forecasting model using arima, generalized autoregressive conditional heteroscedastic and support vector machines. American Journal of Applied Sciences, 11(3), 425. https://doi.org/10.3844/ajassp.2014.425.432 Ayele, A. W., Gabreyohannes, E., & Tesfay, Y. Y. (2017). Macroeconomic determinants of volatility for the gold price in Ethiopia: the application of GARCH and EWMA volatility models. Global Business Review, 18(2), 308-326. Bank of Tanzania (BOT). (2022). Monthly Economic Review. 210 Letema & Mbwambo, International Journal of Research in Business & Social Science 12(1) (2023), 204-211 Cerović Smolović, J., Lipovina-Božović, M., & Vujošević, S. (2017). GARCH models in value at risk estimation: empirical evidence from the Montenegrin stock exchange. Economic research-Ekonomska istraživanja, 30(1), 477-498. https://doi.org/10.1080/1331677X.2017.1305773 Charles, A., & Darné, O., & Charles, A., & Darné, O. (2021). Econometric history of the growth–volatility relationship in the USA. 15. https://doi.org/10.1016/j.compchemeng.2021.107513 Deebom, Z. D., & Essi, I. D. (2017). Modeling price volatility of nigerian crude oil markets using garch model: 1987-2017. International Journal of Applied Science & Mathematical Theory, 4(3), 23. Ekong, C. N., & Onye, K. U. (2017). Application of Garch models to estimate and predict financial volatility of daily stock returns in Nigeria. https://mpra.ub.uni-muenchen.de/88309/ Fondo, K. S., O., A. A. ,. Kiti, L. A., & Otulo, C. W. (2021). Modeling of Petroleum Prices in Kenya Using Autoregressive Integrated Moving Average and Vector Autoregressive Models. 17(6), 18–27. https://doi.org/10.9790/5728-1706011827 Haque, M. I., & Shaik, A. R. (2021). Predicting crude oil prices during a pandemic: A comparison of arima and garch models. Montenegrin Journal of Economics, 17(1), 197–207. https://doi.org/10.14254/1800-5845/2021.17-1.15 Herrera, A. M., Hu, L., & Pastor, D. (2018). Food price volatility and macroeconomic factors: Evidence from GARCH and GARCHX estimates. International Journal of Forecasting, 34, 622-635. JICA. (2022). The Project for Domestic Natural Gas Production and Supply System in Tanzania. https://openjicareport.jica.go.jp/pdf/12342341.pdf Kasongwa, L., & Minja, E. (2022). Moderating effects of oil price on the impact of macroeconomic variables on stock market performance. Business management review, 25(1), 75–89. https://journals.udsm.ac.tz/index.php/bmr/article/download/4936/4204. Khan, M. S., Khan, K. I., Mahmood, S., & Sheeraz, M. (2019). Symmetric and asymmetric volatility clustering Via GARCH family models: An evidence from religion dominant countries. 13(1), 20–25. https://doi.org/10.24312/1900148130104 Liu, L. Q. Geng, Y. Z., & Y., Wang (2022). Investors’ perspective on forecasting crude oil return volatility: Where do we stand today? https://doi.org/10.1016/j.jmse.2021.11.001 Nyongesa, D. N., & Wagala, A. (2016). Nonlinear time series modelling of the diesel prices in kenya. International Journal of Academic Research in Economics and Management Sciences, 5(4). https://doi.org/10.6007/IJAREMS/v5-i4/2368 Rodhan, M., & Jaaz, A. (2022). Box-jenkins modelling and forecasting of wti crude oil price. In Proceedings of 2nd international multi-disciplinary conference theme: Integrated sciences and technologies, imdc-ist 2021, 7-9 september 2021, Sakarya, Turkey. http://dx.doi.org/10.4108/eai.7-9-2021.2314841 Saltik, O., Degirmen, S., & Ural, M. (2016). Volatility modelling in crude oil and natural gas prices. Procedia Economics and Finance, 38, 476-491. Retrieved from https://www.sciencedirect.com/science/article/pii/S2212567116302192 https://doi.org/10.1016/S2212-5671(16)30219-2 Saxena, K., & Ndule, M. A. (2020). The effect of oil price shocks on inflation in tanzania-an autoregressive distributed lag and vector autoregressive approach. Bulletin of Pure & Applied Sciences Mathematics. https://doi.org/10.5958/2320-3226.2020.00025.9 Sekati, B. N. Y., Tsoku, J. T., & Metsileng, L. D. (2020). Modelling the oil price volatility and macroeconomic variables in south africa using the symmetric and asymmet- ric garch models. Cogent Economics & Finance, 8(1), 1792153. Selvi, J., Kaviya Shree, R., & Krishnan, J. (2018). Forecasting crude oil price using arima models. International Journal of Advance Research, 7, 334–343. Shah, J., Kiruthiga, G. (2020). Crude Oil Price Forecasting Using ARIMA model. International Journal of Advanced Scientific Inovation, 1(1), 1–11. Shah P A, Dr. G. K. (2020). Crude Oil Price Forecasting using ARIMA model. Shakiru, T. H., & Liu, X. (2022). The asymmetric relationship between oil price shocks and macroeconomic fluctuation in tanzania: A multivariate time series model approach. https://doi.org/10.22271/maths.2022.v7.i2a.794 Shambulingappa, H., et al. (2020). Crude oil price forecasting using machine learning. International Journal of Advanced Scientific Innovation, 1(1). DOI: https://doi.org/10.5281/zenodo.4641697 Suleiman, S., Alabi, M., Issa, S., Usman, U., & Adamu, U. (2015). Modeling and forecasting the crude oil price in nigeria. International Journal of Novel Research in Marketing Management and Economics, 2(1), 1–13. Wacuka Ng’ang’a, F., & Oleche, M. (2022). Modelling and forecasting of crude oil price volatility comparative analysis of volatility models. Journal of Financial Risk Management, 11(1), 154–187. https://doi.org/10.4236/jfrm.2022.111008 Wei, Y., Wang, Y., & Huang, D. (2010). Forecasting crude oil market volatility: Further evidence using GARCH-class models. Energy Economics, 32(6), 1477-1484. https://doi.org/10.1016/j.eneco.2010.07.0092/13/2023 1:15:00 PM Publisher’s Note: SSBFNET stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. © 2023 by the authors. Licensee SSBFNET, Istanbul, Turkey. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). International Journal of Research in Business and Social Science (2147-4478) by SSBFNET is licensed under a Creative Commons Attribution 4.0 International License. 211