-2

I am building model in R. Can you help me how to interpret this results. For example what is Pr(>|t|) and others...Thanks!

 Call:
lm(formula = Cn ~ B10_75mean + b5_75mean + ui75mean + tvi75mean)

**Residuals:**
    Min      1Q  Median      3Q     Max 
-6.3655 -3.7021 -0.3621  2.9530 12.5155 
**Coefficients:**
             Estimate Std. Error t value Pr(>|t|)
(Intercept)  24.08902  228.38548   0.105    0.917
B10_75mean   -0.04073    0.75413  -0.054    0.957
b5_75mean   -39.98091   42.28717  -0.945    0.355
ui75mean     20.71277   20.59630   1.006    0.326
tvi75mean    32.69517   42.33715   0.772    0.448 
Residual standard error: 4.826 on 22 degrees of freedom
Multiple R-squared:  0.2408,    Adjusted R-squared:  0.1027 
F-statistic: 1.744 on 4 and 22 DF,  p-value: 0.1763
Valeriy
  • 103
  • 4
ABCD
  • 131
  • 5

1 Answers1

1

Pr (>|t|) is the p value. It's actually a better representation than "p-value" because it tries to make you a little more aware of what the p value is (that is, the probability of getting a t that is farther from 0 than the one you got, given a sample the size of the one you have, if the null hypothesis is true).

The rest is really standard output from any regression. Explaining all of the output is too broad a question for this forum and would really require a course in regression. But if you have other specific questions about the output, feel free to ask them.

Peter Flom
  • 94,055
  • 35
  • 143
  • 276
  • Thanks, and is this result saying something bout how close will the predicted and measured values be. – ABCD Oct 12 '15 at 11:27
  • R^2 says something about that, but if that is what you want to find out, you could make a scatterplot of the predicted vs. actual values. – Peter Flom Oct 12 '15 at 11:30