Reputation: 399
I am trying to run a basic time series regression in R.
reg<-lm(y~x)
summary(reg)
Call:lm(formula = y ~ x)
Residuals:
Min 1Q Median 3Q Max
-100.188 -21.600 0.503 21.999 97.296
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 102.53835 4.66296 21.990 <2e-16 ***
x -0.03687 0.04524 -0.815 0.415
However I would also like a visual plot of the y variable, the model's projection and the residuals underneath. Is there a capability in R to have this kind of regression? Thank you!
Upvotes: 0
Views: 272
Reputation: 9167
I'm not totally sure what you want to plot. The following gives you two plots - one with the data and the model prediction and one with the residuals.
par(mfrow=c(2,1)) # Two plots in one window
plot(x,y) # Your datapoints
lines(x,predict(reg)) # The model prediction
plot(x, residuals(reg), ylab='Residuals') # x vs. residuals
R's plotting facilities for linear models are fairly good. I strongly suggest you to have a look at the output of plot(reg)
, which probably gives you far more information than your plots if interpreted correctly.
Upvotes: 2