# Identifying outliers based on standard error of residuals vs sample standard deviation

If this is too elementary a question for CrossValidated and should go elsewhere please let me know.

I've been given a linear model in R similar to:

model = lm(y ~ x1 + x2 + x3, data=sampleset)


...that when fed a set of data and plotted produces the graph below:

The light blue center line represents the linear regression prediction. The blue area represents 2 standard deviations from expected in either direction, using "standard error of residuals":

# se_: number of standard errors away from the predicted value
se_thresh = 2

# standard error of residuals
res_se = sd(model$residuals) # for each member in the sample if(model$fitted.values[j] < (i - se_thresh*res_se)){

-