regressionLectures_4xgs

# regressionLectures_4xgs - Part 1: Linear Regression -...

This preview shows pages 1–5. Sign up to view the full content.

Regression Analysis April 4, 2011 Part 1: Linear Regression - Introduction Example in lab: Design of a casting mold want to reduce the time to Fnish a batch of castings but need to maintain low percentage of defective castings Casting Mold Applications of regression What regression analysis can do: investigate how variables are related test if there is a relationship between variables predict future values of variables control system output

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Fitting equations to data Two types of mathematical models: Theoretical: Derived from Frst principles via mathematical reasoning Empirical: Derived by “Ftting” an equation to data typically the equation has a simple parametric form parameters are estimated by adjusting them to get the best Ft to the data we will study empirical models Linear models (LM’s) E ( Y | X 1 , . . . , X p )= β 0 + β 1 X 1 + β 2 X 2 + · · · + β p X p So Y = β 0 + β 1 X 1 + β 2 X 2 + · · · + β p X p + ǫ where ǫ is variation of Y about E ( Y | X 1 , . . . , X p ) In example: Y is the time to Fnish a cast. X 1 , . . . , X p are the design variables of the mold, e.g., cavity width, riser height, . . . Variables ±rom a previous slide: Y = β 0 + β 1 X 1 + β 2 X 2 + · · · + β p X p + ǫ Y : “outcome”, “response”, “dependent variable” X 1 , . . . , X p : “predictors”, “independent variables”, “covariates” Y , X 1 , . . . , X p : “variables” Parameters ±rom a previous slide: Y = β 0 + β 1 X 1 + β 2 X 2 + · · · + β p X p + ǫ β 0 : “intercept” β 1 , . . . , β p : “regression coe f cients” = “slopes” = “partial derivatives” β j = X j E ( Y | X 1 , . . . , X p ) β 0 , β 1 , . . . , β p : “parameters”
Noise From a previous slide: Y = β 0 + β 1 X 1 + β 2 X 2 + · · · + β p X p + ǫ ǫ is the unpredictable variation in Y ǫ is called the “noise” “error” “residual variation” When is a model linear? From previous page: Y = β 0 + β 1 X 1 + β 2 X 2 + · · · + β p X p + ǫ Defnition: A model is linear Y is linear in the parameters (linear in β 0 , . . . , β p ). Key point: X 1 , . . . , X p can be anything observable, even indirectly. They can be nonlinear ±unctions o± measured quantities. When is a model linear? Which o± these are linear models? Y = β 0 + β 1 × ( CavityWidth )+ β 2 × ( CavityWidth ) 2 , Y = β 0 + β 1 × ( CavityWidth )+( β 2 × CavityWidth ) 2 , Y = β 0 + β 1 × ( CavityWidth exp ( β 2 × CavityWidth ) , Nonlinear models Nonlinear regression is covered in more advanced courses For now, you only need to know a nonlinear model when you see it Example: Y = β 0 + β 1 exp ( β 2 X ǫ This model is nonlinear because Y is a nonlinear ±unction o± β 2

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Data Observe ( Y i , X i , 1 , . . . , X i , p ) , for i = 1 , . . . , n i = index of “observation” = “case” = “subject” = “row in data spreadsheet” So the linear regression model can be rewritten as: Y i = β 0 + β 1 X i , 1 + β 2 X i , 2 + · · · + β p X i , p + ǫ i Notice that β 0 , β 1 , . . . , β p do not depend on i The columns of the data spreadsheet are Y i , X i , 1 , . . . , X i , p Least-squares Estimation The predictor of Y i is h Y i = h β 0 + h β 1 X i , 1 + h β 2 X i , 2 + · · · + h β p X i , p h Y i
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 03/18/2012 for the course ORIE 3120 taught by Professor Jackson during the Spring '09 term at Cornell.

### Page1 / 54

regressionLectures_4xgs - Part 1: Linear Regression -...

This preview shows document pages 1 - 5. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online