Inflation refers to a situation in which the economyâ€™s overall price level is rising and it becomes a major problem in Ethiopia. Additionally, when inflation Volatility rises; it exerts harmful effects on an economy not only through changes in the price level but also through increased price level uncertainty. This study aimed Multivariate Time Series Modeling and Forecasting Inflation Volatility in Ethiopia. To attain the proposed goal, monthly based observations from January 2010 to December 2020 with consumer price index (CPI), food price index (FPI), non-food price index (NFPI) and exchange rates (ER) taken from the National bank of Ethiopia (NBE). STATA 14 version and E-views 9 were used for the purpose of data analysis. In this work, first we compare different Multivariate Generalized Autoregressive Conditional Heteroskedasticity (MGARCH) models, particularly comparing Baba, Engle, Kraft and Kroner (BEKK) - GARCH model with Dynamic Conditional Correlation model (DCC) â€“ GARCH models based on Akaike Information criterion (AIC), Schwartz Bayesian information criterion (SBIC) and Hannan-Quinn information criterion (HQIC). The result of the study has shown that Unit root test reveals that all the series are non-stationary at level and stationary at the log return. The result of the study has also shown that the DCC-GARCH model outperforms in estimating parameters than BEKK-GARCH model. The result of forecasting revealed that high volatility prediction in FPI will increase a high rate, CPI will decrease a low rate, NFPI will decrease for the first twelve months and stable for the rest of twenty four months and ER increase for the first four months and stable for the rest of thirty two months. The DCC- GARCH algorithm is more efficient as compared with BEKK-GARCH algorithm based on time computational complexity. Thus, the moreover, government should attention to reduce inflation by taking action for further interventions.