Spectrum occupancy models can help you make better use of the radio spectrum. It hasrnalso been extensively researched in recent decades as it is critical for developing newrnregulations for spectrum allocation for future technologies as well as monitoring the activitiesrnthat take place on that spectrum. Understanding the amount of available spectrumrnis critical for future wireless technologies that want to address the so-called spectrumrnscarcity issue. Spectrum occupancy measurements provide critical data for frequencyrnplanning and optimization, as well as assist in smart decision-making. The goal of timernseries modeling is to collect and thoroughly examine previous data from a time seriesrnin order to construct an appropriate model that accurately captures the seriesâ€™ intrinsicrnstructure. This thesis examines three types of time series analysis methods: Holt-winters,rnSeasonal Auto Regressive Integrated Moving Average (SARIMA), and SARIMA eXogenousrnregresses (SARIMAX) based models, as well as their inherent prediction strengthsrnand weaknesses. Time series modeling principles such as trend, stationarity, seasonality,rnresidual, and so on have also been covered. To assess the accuracy rate, we fitted multiplernmodels to a time series using five primary metrics. Among the methods used arernmean square error, mean absolute error (MAE), root-mean-square error, mean absoluternpercentage error, and R-squared. At 1800MHz, the maximum spectrum occupancy isrn60.35%, and at 900MHz, it is 44.71%. For 900MHz MAE, the SARIMAX model producedrnbetter predictions (35.34% and 50.1% lower than the SARIMA and Holt-Winterrnmodels, respectively), while for 1800MHz, the SARIMAX model produced 42.6% andrn52.6% lower than the SARIMA and Holt-Winter models, respectively