and your reasoning or justifications. QUESTION1 (22 points) A regression model w

Question

and your reasoning or justifications. QUESTION1 (22 points) A regression model was developed to predict the operating cost of manufacturing plants located in one of the European countries. The data was based on 20 comparable metropolitan cities. The independent variables used were labor cost (X) and electric power cost (X)-The following is the regression model equation. All costs are in millions of Euros per annum p = 0.73 + 0.78 x1 + 1.6 x2 a. Identify the y- intercept and the partial slopes of this model. (3 points) b. Interpret the meaning of the y-intercept and the partial slopes with reference to the context. (6 points) es en os each math 315 iferevl.doc 1215

Explanation / Answer

d) this is due to multi-collinearity

as R^2 increases , standard error also increases

https://www3.nd.edu/~rwilliam/stats1/x91.pdf

e)

for a multiple regression model we plot the residuals given by the model against (1) values of

each independent variable, (2) predicted values of the dependent variable, and (3) the time

order in which the data have been observed (if the regression data are time series data). A

fanning-out pattern on a residual plot indicates an increasing error variance; a funneling-in

pattern indicates a decreasing error variance. Both violate the constant-variance assumption. A

curved pattern on a residual plot indicates that the functional form of the regression model is

incorrect. If the regression data are time series data, a cyclical pattern on the residual plot

versus time suggests positive autocorrelation, while an alternating pattern suggests negative

autocorrelation. Both violate the independence assumption. On the other hand, if all residual

plots have (at least approximately) a horizontal band appearance, then it is reasonable to believe

that the constant-variance, correct functional form, and independence assumptions approxi-

mately hold. To check the normality assumption, we can construct a histogram, stem-and-leaf

display, and normal plot of the residuals. The histogram and stem-and-leaf display should look

bell-shaped and symmetric about 0; the normal plot should have a straight-line appearance.

f)standard error of estimate = sqrt(SSE/(n-k-1))

= sqrt(300/(50 -4-1))

= 2.581988

this is also called MSE , it is used in finding confidence interval of beta coeffecients as well as that of estimate of dependent variable

Please rate

Related Questions

Navigate

• Browse (All)
• Browse A
• Subjects

---

---

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