Modeling - Build the models

 Model Building Process

1. Clean data and understanding of content are crucial.
2. Goals include better predictions, object classification, and system understanding.
3. Focused phase compared to exploratory analysis.
4. Outcomes determined by desired outcomes.
5. Below Figure illustrates model building components.

 Building a model is an iterative process. The way you build your model depends on whether you go with classic statistics or the somewhat more recent machine learning school, and the type of technique you want to use. Either way, most models consist of the following main steps:

 1. Model and variable selection

• Selecting variables and modeling technique based on exploratory analysis findings.
• Judgment required to choose the right model for a problem.
• Consideration of model performance and project requirements.
• Factors to consider: model's suitability for production environment, maintenance challenges, and model's ease of explanation.
• Action required once the model is developed.

2. Model execution

Once you’ve chosen a model you’ll need to implement it in code. Here are the two example 

Example1:

In the above code we provided how a linear regression model will be executed.

Example2:

 3. Model diagnostics and model comparison

• Multiple models are built and chosen based on multiple criteria.
• Holdout sample is used to evaluate the model after building.
• The model should work on unseen data.
• Only a fraction of the data is used for model estimation.
• The model is then unleashed on unseen data and error measures calculated.
• Multiple error measures are available, with the mean square error.