IF4006 Forecasting and Optimization Syllabus:

IF4006 Forecasting and Optimization Syllabus – Anna University PG Syllabus Regulation 2021

COURSE OBJECTIVES:

 Knowledge to build and apply time series forecasting models
 Learn what attributes make data a time series.
 Learn about seasonality, trends, and cyclical patterns.
 Load and Summarize Dataset
 Load and Plot Dataset

UNIT I TIME SERIES FORECASTING

Different types of data. Cross-sectional data. Time series data. Panel data. Internal structures of time series. General trend. Seasonality. Run sequence plot. Seasonal sub series plot. Multiple box plots. Cyclical changes. Unexpected variations. Models for time series analysis. Zero mean models. Random walk. Trend models. Seasonality models. Forecasting Time Series. Estimation of Transfer Functions. Analysis of Effects of Unusual Intervention Events to a System. Analysis of Multivariate Time Series.

UNIT II TIME SERIES DATA PREPARATION

Common Data Preparation Operations for Time Series. Time stamps vs. Periods. Converting to Timestamps. Providing a Format Argument. Indexing. Time/Date Components. Frequency Conversion. Time Series Exploration and Understanding. How to Get Started with Time Series Data Analysis. Data Cleaning of Missing Values in the Time Series. Time Series Data Normalization and Standardization. Time Series Feature Engineering. Date Time Features. Lag Features and Window Features. Rolling Window Statistics. Expanding Window Statistics.

UNIT III LINEAR STATIONARY MODELS

Stochastic Processes and Stationarity. Wold’s Decomposition and Autocorrelation. First-Order Autoregressive Processes. Second-Order Autoregressive Process. First-Order Moving Average Processes. Second-Order Moving Average Process. Estimation of the Partial Autocorrelation Function. Standard Errors of Partial Autocorrelation Estimates. General AR and MA Processes. Autoregressive-Moving Average Models. ARMA Model Building and Estimation. Duality Between Autoregressive and Moving Average Processes.

UNIT IV REGRESSION EXTENSION TECHNIQUES FOR TIME-SERIES DATA

Autoregressive Integrated Moving Average. Seasonal ARIMA. SARIMAX. Vector Auto regression. VARMA. Nonstationary First-Order Autoregressive Process. General Model for a Nonstationary Process Exhibiting. Homogeneity. General Form of the ARIMA Model. Three Explicit Forms for the ARIMA Model. Difference Equation Form of the Model. Random Shock Form of the Model. Inverted Form of the Model. Integrated Moving Average Processes. Integrated Moving Average Process of Order (0, 1, 1). Integrated Moving Average Process of Order (0, 2, 2). Prophet Forecasting.

UNIT V DEEP LEARNING FOR TIME SERIES FORECASTING

Training MLPs. Automatically Learning and Extracting Features from Raw and Imperfect Data. Deep Learning Supports Multiple Inputs and Outputs. MLPs for time series forecasting. Bidirectional recurrent neural networks. Deep recurrent neural networks. Training recurrent neural networks. Solving the long-range dependency problem. Long Short Term Memory. Gated Recurrent Units. Recurrent neural networks for time series forecasting. 2D convolutions. 1D convolution. 1D convolution for time series forecasting.

LIST OF EXPERIMENTS : 30

1: Time Series Prediction of stock prices using ARIMA Model
2: Time Series Prediction of rainfall data using SARIMA Model
3: Forecasting of agricultural commodity pricing using pro
4: Time Series Prediction of Car Sales using ARIMA and SARIMA Model
5: Predicting Air Traffic Flow using Deep Learning

COURSE OUTCOMES:

CO1: Compile and fit time series forecasting model to training data
CO2: Evaluate Forecast Model
CO3: Analysis and compare ARIMA vs SARIMA vs Deep Learning Vs Prophet
CO4: How to evaluate a Prophet model on a hold-out dataset.
CO5: Assess trained model performance

TOTAL : 45+30 =75PERIODS

REFERENCES

1. Machine Learning for Time Series Forecasting with Python, Francesca Lazzeri, PhD. Wiley 2020
2. Practical Time Series Analysis, Dr. Avishek Pal and Dr. PKS Prakash. Packt Publishing, 2017
3. Hands-on Time Series Analysis with Python, B V Vishwas and Ashish Patel. Apress,2020
4. DEEP TIME SERIES FORECASTING With PYTHON, Dr. N.D Lewis,2016
5. Practical Time Series Analysis, Aileen Nielsen. O’Reilly Media, 2019