how to remove values that affect a nonlinear equation by using spss software kno

Question

how to remove values that affect a nonlinear equation by using spss software knowing that y=a*x1+b*x2+c*x3+d*x3+e*x4+f*x4+g knowing that data n=1052 ,show me the way to remove the values that affect the R2 .please i want the method to remove data one by one until it reach the approprate valide sample size, using spss .iasked this question twise and you ask me to give you the data , i cant give you the matrix of data you want because it is huge and you dont accept such data please show me the steps to remove data and what i remove is it the out liers in y or in x1 and x2 or..... x4 .what is the conditions that must be?

Explanation / Answer

Look at the individual error terms. The one with very high error terms(precisely with absolute value higher than 2) can be removed. You can also look at Cook's Distance of individual point. Cook's Distance measures the change in Regression Model after deleting that observation. The points with high Cook's D values are treated as influential. You can delete those values to get good regression model.

Related Questions

Navigate

• Browse (All)
• Browse H
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.