Build practical time series forecasting skills across Excel, R, Python, and EViews.
Turn historical data into reliable forecasts using regression, exponential smoothing, ARIMA, SARIMA, and ARMA models.
This Specialization develops a complete forecasting workflow, from identifying trends and seasonality to building, validating, interpreting, and refining predictive models. You will begin with accessible Excel-based forecasting using weighted averages, exponential averages, correlation, and regression. You will then apply R to decomposition, regression-based forecasting, ACF and PACF diagnostics, and advanced ARIMA and SARIMA modeling.
Using Python, you will preprocess data, handle missing values and outliers, select features, create regression and time series models, evaluate performance, and communicate results. Finally, you will use EViews to estimate and assess univariate ARMA models through correlograms, residual analysis, parameter significance, and the Ljung-Box Q test.
Through hands-on applications in climate analysis, workforce analytics, business, finance, and operations, you will learn to select suitable methods, compare model performance, and produce evidence-based forecasts for informed decision-making.
Applied Learning Project
Complete hands-on projects using climate, workforce, business, and financial datasets to identify trends, seasonality, and relationships across time. Apply Excel, R, Python, and EViews to preprocess data, build and compare regression, ARMA, ARIMA, SARIMA, and smoothing models, validate accuracy, and produce evidence-based forecasts for authentic planning and decision-making problems.
