Forecasting Models with Python

Full Course Content Last Update 06/2018

Learn forecasting models through a practical course with Python programming language using S&P 500® Index ETF prices historical data. It explores main concepts from basic to expert level which can help you achieve better grades, develop your academic career, apply your knowledge at work or do your business forecasting research. All of this while exploring the wisdom of best academics and practitioners in the field.

Become a Forecasting Models Expert in this Practical Course with Python

• Read S&P 500® Index ETF prices data and perform forecasting models operations by installing related packages and running code on Python PyCharm IDE.
• Estimate simple forecasting methods such as arithmetic mean, random walk, seasonal random walk and random walk with drift.
• Evaluate simple forecasting methods forecasting accuracy through mean absolute error and root mean squared error scale-dependent metrics.
• Approximate simple moving averages and exponential smoothing methods with no trend or seasonal patterns such as Brown simple exponential smoothing method.
• Estimate exponential smoothing methods with only trend patterns such as Holt linear trend, exponential trend, Gardner additive damped trend and Taylor multiplicative damped trend methods.
• Approximate exponential smoothing methods with trend and seasonal patters such as Holt-Winters additive seasonality and Holt-Winters multiplicative seasonality methods.
• Select exponential smoothing method with lowest Akaike and Schwarz Bayesian information loss criteria.
• Asses simple moving average and exponential smoothing methods forecasting accuracy through mean absolute error and root mean squared error scale-dependent metrics.
• Identify Box-Jenkins autoregressive integrated moving average model integration order through level and differentiated first order trend stationary time series augmented Dickey-Fuller unit root test.
• Recognize autoregressive integrated moving average model autoregressive and moving average orders through autocorrelation and partial autocorrelation functions.
• Estimate non-seasonal autoregressive integrated moving average models such as random walk with drift, differentiated first order autoregressive, Brown simple exponential smoothing, Holt linear trend and Gardner additive damped trend models.
• Approximate seasonal autoregressive integrated moving average models such as seasonal random walk with drift, seasonally differentiated first order autoregressive and Holt-Winters additive seasonality models.
• Choose autoregressive integrated moving average model with lowest Akaike and Schwarz Bayesian information loss criteria.
• Evaluate autoregressive integrated moving average models forecasting accuracy through mean absolute error and root mean squared error scale-dependent metrics.
• Assess highest forecasting accuracy autoregressive integrated moving average model residuals or forecasting errors white noise requirement through Ljung-Box lagged autocorrelation test.

Become a Forecasting Models Expert and Put Your Knowledge in Practice

Learning forecasting models is indispensable for business or financial data science applications in areas such as sales and financial forecasting, inventory optimization, demand and operations planning, and cash flow management. It is also essential for academic careers in data science, applied statistics, operations research, economics, econometrics and quantitative finance. And it’s necessary for business forecasting research.

But as learning curve can become steep as complexity grows, this course helps by leading you step by step using S&P 500® Index ETF prices historical data for forecast modelling to achieve greater effectiveness. 

Content and Overview

This practical course contains 41 lectures and 5.5 hours of content. It’s designed for all forecasting models knowledge levels and a basic understanding of Python programming language is useful but not required.

At first, you’ll learn how to read S&P 500® Index ETF prices historical data to perform forecasting models operations by installing related packages and running code on Python PyCharm IDE.

Then, you’ll define simple forecasting methods such as arithmetic mean, random walk, seasonal random walk and random walk with drift. Next, you’ll evaluate simple methods forecasting accuracy through mean absolute error and root mean squared error scale-dependent metrics. 

Next, you’ll define simple moving averages and exponential smoothing methods. For exponential smoothing methods with no trend or seasonal patterns, you’ll define Brown simple exponential smoothing method. For exponential smoothing methods with only trend patterns, you’ll define Holt linear trend, exponential trend, Gardner additive damped trend and Taylor multiplicative damped trend methods. For exponential smoothing methods with trend and seasonal patterns, you’ll define Holt-Winters additive seasonality and Holt-Winters multiplicative seasonality methods. After that, you’ll select exponential smoothing method with lowest Akaike and Schwarz Bayesian information loss criteria. Later, you’ll evaluate simple moving average and exponential smoothing methods forecasting accuracy through mean absolute error and root mean squared error scale-dependent metrics.

After that, you’ll define Box-Jenkins autoregressive integrated moving average models. Then, you’ll identify autoregressive integrated moving average model integration order through level and differentiated time series first order trend stationary augmented Dickey-Fuller unit root test. Next, you’ll identify autoregressive integrated moving average model autoregressive and moving average orders through autocorrelation and partial autocorrelation functions. For non-seasonal autoregressive integrated moving average models, you’ll define random walk with drift, differentiated first order autoregressive, Brown simple exponential smoothing, Holt linear trend and Gardner additive damped trend models. For seasonal autoregressive integrated moving average models, you’ll define seasonal random walk with drift, seasonally differentiated first order autoregressive and Holt-Winters additive seasonality models. After that, you’ll select autoregressive integrated moving average model with lowest Akaike and Schwarz Bayesian information loss criteria.  Later, you’ll evaluate models forecasting accuracy through mean absolute error and root mean squared error scale-dependent metrics. Finally, you’ll assess highest forecasting accuracy autoregressive integrated moving average model residuals or forecasting errors white noise requirement through Ljung-Box lagged autocorrelation test.