-
In artificial_data.ipynb, we generate four different types of seasonality and describe each one to understand how these seasonal patterns behave.
-
In descriptive-data.ipynb, we use the ydata_profiling package to quickly generate a profile report.
-
In Time-series-data.ipynb, we describe basic statistics from our dataset that will be used in the next Jupyter notebook.
-
In Cross-correlation.ipynb, we use the Cross-Correlation Function (CCF) to examine the correlation between each variable and the target variable at lag 0, and to assess the correlation at further lags.
-
In Studying-target.ipynb, we conducted our first modeling using ETL models.
-
In Studying-target-2.ipynb, we create a new dataset using a pipeline. In addition, we perform an XGBoost model to forecast and then analyze the results. I will try another approach in the future, which will also allow the use of confidence intervals.