A toy example to demonstrate MLR for relationship between ancestry and age of cancer diagnosis

See: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4069235/

For: http://stats.stackexchange.com/questions/173708/assessing-the-relationship-between-continuous-variables


In [4]:
data <- read.csv('173708.csv', header=T)

In [5]:
data


Out[5]:
Age.of.DiagnosisNative.AmericanEuropeanAfrican
1550.4780950.4643910.057514
2590.2532850.1457520.600963
3580.5546350.4117770.033588
4440.4445490.5371770.018274
5530.5107560.4791370.010106
6520.4528830.3560770.19104

In [6]:
fit <- lm(Age.of.Diagnosis ~ Native.American +  European + African, data=data)

In [7]:
print(summary(fit))


Call:
lm(formula = Age.of.Diagnosis ~ Native.American + European + 
    African, data = data)

Residuals:
         1          2          3          4          5          6 
 3.944e+00  8.503e-01  5.642e-01 -1.545e+00  1.657e-09 -3.814e+00 

Coefficients:
                Estimate Std. Error t value Pr(>|t|)
(Intercept)      1252532    4707189   0.266    0.815
Native.American -1252432    4707191  -0.266    0.815
European        -1252532    4707189  -0.266    0.815
African         -1252478    4707187  -0.266    0.815

Residual standard error: 4.094 on 2 degrees of freedom
Multiple R-squared:  0.7696,	Adjusted R-squared:  0.4239 
F-statistic: 2.226 on 3 and 2 DF,  p-value: 0.3249


In [8]:
layout(matrix(c(1,2,3,4),2,2))
plot(fit)


Warning message:
: not plotting observations with leverage one:
  5Warning message:
: not plotting observations with leverage one:
  5