Assignment
On Blackboard/Assignments/, you will find a data file named "bankwire.txt", which shows daily volume of wire transfers between Mar 1st 2014 and Sep 16th 2014.
1. Provide a time series plot (a line plot) of the data. Is there a trend or a seasonal factor?
2. Compute the sample autocorrelations and partial autocorrelations (Correlogram) up to displacement of 40 (default). Provide the graphs of autocorrelations and partial autocorrelations with a 95% confidence interval.
3. Diagnostic Check: Suppose that,based on the correlogram in #2, you narrow down to5 candidates, ARMA (2,0), ARMA (2,2), ARMA (2,1), ARMA (1,1), and ARMA (1,2), for your final forecasting model.
(1) Attach the regression outputs of the 5 models.
(2) Calculate the AIC and BIC values for each model.
(3) Provide the Residual sample autocorrelation and partial autocorrelation functions for each model.
4. Select a parsimonious ARMA (p,q) model that fist the transfer data best. Defend your selection.
5. Forecasting: Use the model selected in 5 to produce forecasts for 5 days (9/17, 9/18, 9/19, 9/20, and 9/21).