[R] Predict

Ricardo L Gómez rgomezye at educ.umass.edu
Thu Dec 18 00:47:22 CET 2008


Dear Cristopher,

I have a question and hope you may help me...I designed a model with a 
data set, the model

Call:
lm(formula = outcom ~ vari1 + vari2 + dummy1 + dummy2)

Residuals:
Min 1Q Median 3Q Max
-0.7090 -0.2648 0.1466 0.2167 0.7513

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.528295 0.205280 7.445 4.79e-07 ***
vari1 0.067471 0.042248 1.597 0.12676
vari2 0.009433 0.040919 0.231 0.82015
dummy1 1.275861 0.256066 4.983 8.27e-05 ***
dummy2 1.326405 0.414477 3.200 0.00471 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.458 on 19 degrees of freedom
Multiple R-squared: 0.9772, Adjusted R-squared: 0.9724
F-statistic: 203.8 on 4 and 19 DF, p-value: 2.564e-15

but now, using this model, I need to calculate the predicted value of 
OUTCOM (only for one observation) , when vari1=8 and vari 2 =64, and 
also the confidence interval....(the data set has total of 24 
observations),

-- 
Ricardo Gómez 
Center for International Education
University of Massachusetts-Amherst, School of Education
http://www.umass.edu/cie/

**************************************

Scoff at all knowledge and despise
Reason and Science, those flowers of mankind. 
Let the father of all lies
With dazzling necromancy make you blind, 
Then I'll have you unconditionally.
(Faust, Goethe)



More information about the R-help mailing list