This paper proposes a method that utilizes machine learning techniques to predict rental car prices in the metaverse. Theprimary objective of this paper is to predict rental car prices in a metaverse environment. To achieve this goal, we introduce aprediction a method based on actual data provided by KAFLIX. Three experiments were conducted in total, employing variousmachine learning methods including regression analysis and ensemble techniques. The aim was to investigate factors influencingrental car prices, encompassing single-step forecasting, multi-step forecasting, and multi-step forecasting with application of BOHB,an optimization algorithm, to enhance the accuracy of rental car price predictions. As a result of the experiment, it was concludedthat the previous rental car price and date information play crucial roles in predicting rental car prices. Simultaneously, theeffectiveness of utilizing this information for rental car price prediction was demonstrated. In the event that the rental car businessin the metaverse undergoes a development similar to that in reality, our research results are anticipated to offer valuable insightsinto the rental car industry within this enviornment.