I am doing a time series regression between 2 variables. I used the dynlm library in R. I'm trying to understand how to interpret the results.
Could you please point out where I am getting it wrong:
1) The R squared seems very low -- this indicates a weak linear relationship. Or too many outliers. 2) Normal QQ plot shows that there are a significant number of outliers -- at a later date in the time series? 3) Does the 'Residuals vs Fitted' plot show that the model is a pretty good fit for most of the data, except for those outliers?
Any suggested readings (esp open-sourced materials available online) also appreciated.
Call: dynlm(formula = P ~ L(B)) Residuals: Min 1Q Median 3Q Max -63.711 -27.687 -14.907 2.364 146.157 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 17.485051 13.500833 1.295 0.197 L(B) 0.019422 0.002384 8.146 5.84e-14 --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 Residual standard error: 48.17 on 182 degrees of freedom Multiple R-squared: 0.2672, Adjusted R-squared: 0.2632 F-statistic: 66.36 on 1 and 182 DF, p-value: 5.838e-14