smb-1.NW7.20.blank.12.31.03

# smb-1.NW7.20.blank.12.31.03 - (N&W 7.20 EXECUTIVE...

This preview shows pages 1–2. Sign up to view the full content.

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

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: (N&W 7.20) EXECUTIVE SUMMARY (12.31.03) Problem : Data for ___ intermediate and senior level research mathematicians has been provided. Our object is to characterize the dependent variable SALARY (S) (in \$1,000). Along with the value of SALARY for the ___individuals, we have been provided with the values of three independent variables: publication quality PUBQUAL (P), years of experience YREXPER (Y) and grant support GRANTSUP (G). Besides characterizing the relationship between these four variables, we wish to predict SALARY for an individual with PUBQUAL=5, YREXPER=10, GRANTSUP=6 and give a 95% confidence interval for these predictions. Variable Mean Std.Dev. Min Max Shape of Distribution SALARY _____ _____ _____ _____ ____________________ PUBQUAL _____ _____ _____ _____ ____________________ YREXPER _____ _____ _____ _____ ____________________ GRANTSUP _____ _____ _____ _____ ____________________ Recommended Model: =______ + ______*(P-5) + ______*(Y-25) + ______*(G-6) S ˆ Salary (S) in units of ______; Publication Quality (P), Years of Experience (Y), Grant Support (G) s S.P,Y,G =______ (in ______), R 2 =______; Using this first order model, for an individual with PUBQUAL=5, YREXPER=10, GRANTSUP=6, we would predict a SALARY of ______ (in ______) with 95% confidence interval for this individual prediction given by the interval (______,______), (in ______). ------------------------------------ (1) Data checked for accurate input into computer. (2) Assumptions for first order model: mean, variance and normality of residuals ok, although some concern for lower variability for low and high predicted values. Test for independence not possible. (3) β 1 , β 2 , β 3 ,. all significantly different from 0 (p<0.004); F=____, p=______ for comparison with mean of Salary model (4) second order model not significantly better than first order model, F=______, p>_____ (5) =______ + ______*P * + ______*Y * +______*G * with S ˆ P * =(P-3.1)/(8-3.1), Y * =(Y-5)/(47-5), G * =(G-3.5)/(8.3-3.5) (6) =______ + ______*P + ______*Y +______*G original model from computer. S ˆ You may give only this model in your report and omit footnotes (5) and (6)----------------------------------------------------------------------------------------------------------------- ANALYSIS (written before Executive Summary) Problem : Data for ______ intermediate and senior level research mathematicians has been provided. Data for ______ intermediate and senior level research mathematicians has been provided....
View Full Document

{[ snackBarMessage ]}

### Page1 / 2

smb-1.NW7.20.blank.12.31.03 - (N&W 7.20 EXECUTIVE...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online