STAT Linear Regression
The final address charge be no added than 8 pages. SAS cipher acclimated to
generate the after-effects (plots and tables) charge be submitted, organized
into an addendum to the final report. The 8 folio absolute doesn't
include the SAS cipher appendix.
Discussion is allowed, but you charge address the report
independently. Please accredit to the Marfan Syndrome Case Study as
a acceptable archetype of abstruse writing.
TA Yucong Zhang, [email protected] will authority appointment hours on Wed
1-3pm in MATH 541 if you accept activity accompanying and/or SAS coding
questions. I will acknowledgment questions afterwards the class.
You can acquisition out added about SAS cipher from SAS abutment website:
***** SAS command to apprehend in data:
infile 'data-proj.csv' dlm=",";
input FFMC DMC DC ISI acting RH wind rain area;
proc book data=proj;
***** Dataset Description:
Number of Instances: 517
Missing Attribute Values: None
1. FFMC - FFMC basis from the FWI system: 18.7 to 96.20
2. DMC - DMC basis from the FWI system: 1.1 to 291.3
3. DC - DC basis from the FWI system: 7.9 to 860.6
4. ISI - ISI basis from the FWI system: 0.0 to 56.10
5. acting - temperature in Celsius degrees: 2.2 to 33.30
6. RH - about clamminess in %: 15.0 to 100
7. wind - wind acceleration in km/h: 0.40 to 9.40
8. rain - alfresco rain in mm/m2 : 0.0 to 6.4
9. breadth - the austere breadth of the backwoods (in ha): 0.00 to 1090.84
***** Potential Tasks:
1. Acknowledgment capricious is ln(area+1). Create a new cavalcade in
the csv book for the response. Save a archetype of the csv book with
10 variables. This is the abstracts book absolutely acclimated for the project.
2. Accomplish besprinkle plots and accept arbitrary statistics application proc
univariate. Examine the dataset (all 10 variables -- including the 9
original variables additional the acknowledgment variable).
3.1. Detect outliers application the measures alien in
class. Discuss whether you adjudge to accommodate or exclude certain
observations from the study
3.2. Perform archetypal diagnostics, analysis residuals. How do you fix the
problems and access at your final model?
3.3. Perform capricious alternative to acquisition the best model
Note assignment 3.1-3.2-3.3 may charge to be run iteratively, and may not
be in this order, to acquisition the best result.
4. Besides address about the action of applicable the regression
model, address your final R^2, AND artifice predicted breadth with the
observed breadth in the dataset to appearance how able-bodied your archetypal fits the
data. The artifice is the one from SAS proc reg, with ln(area+1) as
the response. (You don't charge to accomplish the artifice on the aboriginal
scale for area).
Order a unique copy of this paper