The result in problem one generalizes:we can consider any regression of Y_i on X

Question

The result in problem one generalizes:we can consider any regression of Y_i on X_i,...,X_ik with an intercept as being the same as a regression of y_i on x_i1,....,x_ik (differenced from means) without an intercept. Using this simplification, consider the regression of y_i on x_i1 and x_i2 with no intercept.

(a)Use the normal equation from the regression of y_i on x_i1 and x_i2 to derive an expression for the slope coefficents in terms of b1=(Sum y_i* x_i1)/(Sum(x_i1)^2) and b2==(Sum y_i* x_i1)/(Sum(x_i1)^2) the lope coefficient from the two simple regression of y_i on x_i1 and y_i on x_i2.

(b)in part (a) you should have terms like =(Sum x_i1* x_i2)/(Sum(x_i1)^2) and =(Sum x_i1* x_i2)/(Sum(x_i2)^2). give an interpretation of these terms.

(c)what does would it mean if the terms in part(b) were zero? what would the relationship between the slope coefficients from the long regression in (a) and the term b1 and b2 look like? how would you interpret this case?

Explanation / Answer

http://criticalman.com/forex/forex_indicator_regression.html Hope this link helps...thanks

Related Questions

Navigate

• Browse (All)
• Browse T
• Subjects

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