Hyderabad is highly affected by climate change and is reported to be a highly prone skin disease-endemic area. This study investigates the association between skin diseases and climate factors. For selecting the best-fit climate prediction method for skin diseases occurrence in Hyderabad city, we have considered 3 different machine learning regression models namely: Poisson Distributed Lag Model [PDLM], Seasonal Autoregressive Integrated Moving Average Model [SARIMA] and other model Standard Multiple Regression [SMR] directed toward investigate the relationship between skin diseases and climate attributes incident during the time period 2000 to 2018. We verify the models lag predicting skin diseases for the time duration of January month to December month 2019 using the MAPE. ROC curves were considered to examine the prediction rate of a skin disease outburst. The results show that relative humidity and temperature are significant parameters which promote skin diseases where there is no rainfall effect. The PDLM model presents the finest fitting prediction of skin disease occurrence and identification of an outbreak when analyzed for a 6, 9, and 12 month time. Nevertheless, the SARIMA algorithm enacts a better prediction of skin disease occurrence for a short 3 month time period. The standard multiple regressions present a high loss prediction of skin disease incidence. From our results we are encouraged to carry out an extensive analysis to validate and examine the modelwith more data in Hyderabad city and contribute in prevention and control of skin diseases at an early stage.