Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: THEend8_
ECON 5248
Please hand in this problem set by March 8, 5:00 pm on Turnitin. The
work is to be your own. Please use Rmarkdown to write your report.
Q1. Obtain the latest daily exchange rate for AUD to USD for a year as
a csv file and then read them into R. Call all the exchange rates you have
except for the last 15, the training sample with the last 15 the holdout
sample.
a. Use the training sample of exchange rates to fit a NAIVE predictive
model. Please report the model fitted using statistics such as the stan-
dard deviation of the residuals, time plot of the standardized residuals,
as well as other statistics that you think are appropriate.
b. Determine how well the model does by checking:
i. whether the residuals are white noise.
ii. whether the residuals are approximately normally distributed us-
ing a normal quantile plot.
c. Predict the holdout sample of 15 exchange rates. i.e., those observa-
tions that are NOT in the sample.
1
d. Plot the actual and predicted exchange rates, as well as the 95% pre-
diction intervals for the actual exchange rates.
Discuss the plot.
e. Compute the root mean squared prediction error for the holdout sam-
ple; i.e.,
RMSPE =
( 15∑
i=1
(yi − ŷi)2
) 1
2
,
where yi is the exchange rate and ŷi is the prediction.