Learning path
Time Series Forecasting
Trend and seasonality, stationarity and differencing, ARIMA and honest forecast evaluation, with animated films and runnable Python. Part of the Data Scientist premium track.
-
1
Time Series: Trend, Seasonality and Moving Averages
Break any time series into trend, seasonality and noise, smooth it with moving averages, and make your first simple forecasts with pandas.
-
2
Stationarity, Differencing and Autocorrelation
What stationarity means, why forecasting models need it, how differencing and log transforms create it, and how to read autocorrelation (ACF) to spot patterns and choose lags.
-
3
ARIMA Forecasting in Python
Understand AR, I and MA, fit ARIMA and seasonal SARIMA models with statsmodels, forecast with confidence intervals, and evaluate on a time-based test set, plus a from-scratch AR…