2. [20 points] Prediction error
You just developed a neural network that predicts the stock price that Google will have at the
end of the day tomorrow. Testing your algorithm on a historic dataset, you find that the error in
your prediction (often called the model \"residuals\" ) can be modeled with a normal distribution
with a mean value of 0.5 dollars, and a standard deviation of 1.2 dollars.
(a) What is the probability that your model prediction will be more than 2 dollars greater than
the actual stock price at the end of the day tomorrow?
(b) You run your model 10 days in a row, without making any changes to it. What is the
probability that at least one of your predictions will have a residual above the 95th percentile
value?
