Forecasting Monthly Sales of French Champagne - Perrin Freres

Trabajo Presentado en el Máster de Big Data, Data Science e IA del tema de Series Temporales

Need to predict how many passengers are going to opt for the airline base on the historical information provided by the Airlines. Using various Time series techniques predicted the number of passengers

LSTM analysis including its helper functions, Pandas Profiling, plotting of the time series, Exponential Smoothing, Simple Exp Smoothing, Holt, Augmented Dickey Fuller test.

Splitting data, Moving Average, Time series decomposition plot, ACF plots and PACF plots, Evaluation Metric MAPE, Simple Exponential Method, Holt method, Holts winter exponential smoothing with additive seasonality and additive trend, Holts winter exponential smoothing with multiplicative seasonality and additive trend, Final Model by combining …

I implement a ETS (ANN) and ETS (AAA) model, followed by a Simple Exponential Smoothing, Holt and Holt-Winters model. In conclusion I compare the results and recommend the best alternative.