Data Science - Time Series Forecasting with Facebook Prophet in Python - Prophet in Code: Holidays and Exogenous Regress

The lecture continues with building profit models for time series, focusing on incorporating holidays and exogenous regressors. It demonstrates creating a new profit model, M3, and adding country holidays for demonstration purposes. The lecture then explains adding exogenous regressors, including state and school holidays, and creating a new model, M4. The process involves dummy encoding categorical data and fitting the model with added regressors. The lecture concludes with forecasting and analyzing results, highlighting the importance of removing non-relevant days from the time series.

What is the primary focus of the lecture in building profit models?

• Improving computational efficiency

• Developing new algorithms

• Enhancing data visualization

• Incorporating holidays and exogenous regressors

Which country is used as an example for adding holidays in the profit model?

• France

• Germany

• United States

• Italy

What is the purpose of plotting components in the model?

• To visualize the influence of each holiday

• To compare different models

• To reduce computational time

• To enhance data accuracy

What does the state holiday column in the dataset represent?

• Customer demographics

• Sales data

• Store opening hours

• Different types of holidays

Why is dummy encoding used for the state holiday column?

• To reduce data size

• To enhance data security

• To simplify data entry

• To convert categorical data into numerical format

What is the name of the new model created after adding regressors?

• M4

• M3

• M2

• M1

What is the default mode for incorporating regressors into the model?

• Multiplicative

• Additive

• Seasonality

• Linear

What challenge is highlighted when predicting test set values?

• Lack of true target values

• Insufficient computational resources

• Complexity of algorithms

• Inaccurate data collection

What assumption is made about store operations on Sundays?

• Stores are open all day

• Stores have reduced hours

• Stores are closed

• Stores have special promotions

What is suggested as a better approach for handling days when the store is closed?

• Include them in the time series

• Remove them from the time series

• Predict them separately

• Ignore them completely