Modeling and evaluation

For the modeling and evaluation step, we'll focus on three tasks. The first is to produce a univariate forecast model applied to just the surface temperature. The second is to develop a vector autoregression model of the surface temperature and CO2 levels, using that output to inform our work on whether CO2 levels Granger-cause the surface temperature anomalies.

Univariate time series forecasting

With this task, the objective is to produce a univariate forecast for the surface temperature, focusing on choosing either an exponential smoothing model, an ARIMA model, or an ensemble of methods, including a neural net. We'll train the models and determine their predictive accuracy on an out-of...