Purpose: Aedes mosquito is responsible for transmitting dengue virus. Mosquito lifecycle is known to be influenced by temperature, rainfall and relative humidity. This retrospective study was planned to investigate if climatic factors can be used to predict occurrence of dengue cases in our area. Methods: Number of monthly dengue cases reported for a period of 19 years was obtained. Monthly data of rainfall, temperature and humidity collected from local Weather Station was correlated with number of monthly reported dengue cases. One-way analysis of variance was used to analyze if the climatic parameters differed significantly between seasons. Four prediction models were developed using Negative Binomial Generalized Linear Model analysis. Monthly rainfall, temperature and humidity were used as independent variables and number of dengue cases reported monthly was used as dependent variable. First model considered data of the present month, while next 3 models involved incorporating the data of 1, 2 and 3 months lag phase respectively. Results: Maximum number of cases were reported during post- monsoon period each year. Temperature, rainfall and humidity varied significantly over pre-monsoon, monsoon and post-monsoon periods. Best correlation between the 3 climatic factors and dengue occurrence was at time lag of 2 months. Model-3 involving 2 months lag phase was found to be the best model. Conclusions: This study indicates that temperature, rainfall and relative humidity significantly affect dengue occurrence in our area. The weather-based dengue prediction model can forecast potential outbreak 2 months in advance providing an early warning system for intensifying dengue control measures.