All phenomenon in the real world are nondeterministic. In statistics, modeling is considered as a tool to develop a statistical equation to explain the variability of an observed response variable using one or more explanatory variables. The basic concept of modelling is:

Observed response variable = model + error.

Thus, the model should have a systematic pattern while the error should not have a systematic pattern. Such modeling is generally carried out under regression environment.   The regression models are

very popular among statistics users. The regression models are the key tools in predictive analytics in data science and such models are used when we need to incorporate uncertainty explicitly in the underlying observed data.

Regression analysis is a form of predictive modelling technique which investigates the relationship between a dependent (target) and independent variable (s) (predictor).  Such modelling is useful in data driven decision making (DDDM).

Developing regression models using SPSS is not a difficult task.  Whatever you input to the computer you will get something, “Garbage in garbage out”. The important thing is how to decide the correct model, how it is convey to any users and how to interpret results so that it can be understood by any client for the use of planning based on data.

Learning Objectives

On the successful completion of the course participants will be able to:

• Understand what regression is and how to visualize data,
• Decide the best type of models to be fitted for data
• Validate models
• Derive inferences and interpret results based on SPSS outputs
• Use SPSS with confident for modeling
• Write statistical reports