Abstract Infant mortality has been a major public health issue in Ethiopia. The main objective of this study was to determine socioeconomic, demographic and environmental factors that may have impact on infant mortality in Ethiopia. The study used data from the 2016 Ethiopian Demographic and Health Survey. Survival analysis techniques such as the Kaplan-Meier survival estimates, the Cox proportional hazards model, accelerated failure time model and parametric shared frailty models were employed to analyze the data for survival time of infants. Results showed that out of 2621 children under the age of one year who were born during the five years preceding the date of the survey only 496 experienced the event or died before age one. The Weibull-Gamma shared frailty model was found to be the best model for predicting survival time of infant in Ethiopia. The unobserved heterogeneity in the population of clusters (regions) as estimated by the Weibull-gamma shared frailty model was θ=0.096 (p-value=0.004) indicating the presence of regional variation on infant mortality. Mothers’ educational level, breastfeeding status, household size, birth type, mother occupation, age at first birth and preceding birth interval were significant predictors of the survival time of infants in Ethiopia. Mothers educational level prolonged the survival time of infants by a factor of (𝜙 =1.403 and 𝜙 =2.094) for primary and secondary and above education as compared to uneducated mothers. Educating mothers, breastfeeding and child spacing need to be considered when planning and developing policies against infant mortality.