Paper Title
Time Series Prediction in Retail Sales for Pharmaceutical Products

Abstract
The purpose of this research is to find the best machine-learning model to predict the pharmaceutical sales in the area of Saskatoon Canada. Throughout the study Univariate and multivariate time series models such as ARIMA,LSTM, XGBoostbeen usedwith feature engineering aiming to compare each of the algorithm’s performances to identify best accurate model. The results shows XGBoostmodel has the higher prediction accuracy for the identified feature set. Keywords - ARIMA, LSTM, XG Boost, RMSE, MAPE, R2 Sales Forecasting