Dengue fever is a prominent mosquito-borne viral disease in the tropics that is estimated to infect as many as 400 million people per year. Dengue is endemic to Colombia, South America and it is crucial to be able to predict when outbreaks may occur so that preventative measures may be taken. The primary vector for the virus, the Aedes aegypti mosquito, requires warm temperatures and standing water to live, breed and incubate the virus. Therefore, weather variables such as temperature and precipitation correlate to dengue incidence and can be used to predict the timing and location of dengue outbreaks. While most of the previous research on this topic has focused on temporal prediction of dengue outbreaks in a small area, a spatiotemporal prediction model for the entire country of Colombia was developed for this study using correlations between weather variables and dengue fever incidence data. Temperature and precipitation data from weather stations across Colombia was interpolated via Inverse Distance Weighting, Kriging and Cokriging. Then a prediction model based on the auto-regressive moving average model was developed to compare dengue incidence to each weather variable and to itself at different time lags then to predict future dengue incidence. The accuracy of the prediction model depended on which variables were incorporated into the model, but the most accurate model was the model that only took historical dengue incidence into account. The model performed better in cities than over the country as a whole, which is notable because the majority of cases occur in highly populated areas. Model prediction errors were high, and it is important to note that socio-economic factors, as well as environmental factors, need to be taken into account to create an accurate prediction model.