About Forecasting Certification Training
Early knowledge is the wealth, even if that knowledge is bit imperfect!!! Wouldn’t you want to unlock the mystery of predicting the stock market? And many of us want to understand how companies are managing their inventory and other resources by forecasting their sales.
Here is the solution in the form forecasting technique also called as time series analysis. Forecasting techniques will be applied for time series data. Forecasting Analytics is considered as one of the major branches in big data analytics.
Managers often have to take decisions in uncertain environment and often find themselves in a bad situation due to lack of skills on applying the right analytical techniques on the data. Forecasting techniques helps companies save millions of dollars by adjusting their production schedules and other plans. Forecasting techniques on univariate and multivariate time series analysishave huge applications across the industries and areas such as Operations management, Finance & Risk management, Retails, Telecom and manufacturing.
Moving Averages and smoothing methods, Box- Jenkins (ARIMA) methodology, Regression with time series data, Holts-Winter, Arch-Garch and Neural Network are the methods widely used for forecasting. Arch-Garch and Neural Networks are the advanced techniques in the forecasting analytics which will be used to model the high frequency data such as stock market and big data.
- Electricity usage pattern over a period of years in a region
- Sales of a product over several years
- Stock market data
Things You Will Learn…
Introduction to Forecasting
- Forecasting & its need
- Types of forecasting
- Steps involved in forecasting
- Types of plots – Scatter plot, Time plot, Lag plot, ACF plot
- Autocorrelation & standard error
- Common pitfalls of plots & Aspect ratio
- Time series components – Trend, Cyclical, Seasonal, Irregular
- Ljung box test for identifying randomness
- Forecasting error & the measures associated with it
- Mean Error
- Mean Absolute Deviation
- Mean Squared Error
- Root Mean Squared Error
- Mean Percentage Error
- Mean Absolute Percentage Error
- Forecasting methods based on smoothing
- Moving Average
- Exponential Smoothing
- Decomposition of time series into 4 components
- Additive model
- Multiplicative model
- Mixed model
- Curve fitting – Least square method
- Simple exponential smoothing (SES)
- Forecasting strategy – Separate, Forecast, Combine
- Moving averages
- Naive model
- Naive Trend model
- Simple average model
- Moving average over k time periods
- Exponential smoothing
- Simple exponential smoothing
- Holt’s version
- Winter’s modification
Modeling different components
- Modeling random component
- Models for stationary time series
- Autoregressive model (AR)
- Moving average model (MA)
- Autoregressive Moving Average (ARMA) model
- Autoregressive Integrated Moving average (ARIMA) model
- Building seasonality into ARIMA models
- Simple Linear, Multiple, Weighted regression
- Non-linearity detection
- Scatter plot
- Partial residual plot
- Partial regression plot
- Non-normality detection
- Normal plot
- Jarque-Bera Normality test
- Growth curve – Trend, Linear, Quadratic, Exponential, Sigmoid
- ARCH & GARCH models
Forecasting steps involves:
Data manipulation and cleaning
• Problem formulation and data collection
• Model building and evaluation
• Model implementation to generate forecast
• Forecast evaluation
Tools You Will Learn…
- R – Revolution Analytics is recently acquired by Microsoft but still remains to be an open source software
Forecasting Course Introduction Video
- Forecasting is predicting the future by considering the historical past data. For e.g., companies forecast sales of next quarter by looking into sales of previous quarters. However, data should be in time-series for forecasting the future events.
- Data arranged in a sequence in an order based on time. For e.g., company sales should be arranged in a time sequence (Jan ‘15, Feb ‘15, March ‘15, April ‘15, May ‘15, June ‘15, July ’15) before forecasting the sales of Aug ’15.
- Forecasting is used across all industries & sectors. Majorly it is applicable for Financial services & insurance, Retail & in weather forecasting.
- A lot of tools are used including R, SAS, STATA, MATLAB, Minitab, Excel etc. We at ExcelR teach you forecasting on R which is highly in demand.
- The detailed course outline is provided on the website. All the forecasting models including AR, MA, ARMA, ARIMA, ARCH & GARCH models are taught.
- Yes, forecasting techniques are imperative for one to be a successful data scientist.