首页 > > 详细

STA457 Assignment

 STA457 Assignment 

Instructions: 
1. Download the "British Columbia" Google Flu Trends and Flu Test data using 
GFluTrends<-read_excel("case_study_1_fluwatch.xlsx",sheet="Google Flu 
Trends", skip = 1) 
fluWatch<-read_excel("case_study_1_fluwatch.xlsx", sheet="FluWatch-BC", 
skip = 2) 
tim<-timeSequence(from = "2003-09-28", to = "2015-08-09", by = "week") 
tim1<-timeSequence(from = "2003-09-07", to = "2015-08-23", by = "week") 
GFT<- timeSeries(GFluTrends[,"British Columbia"], charvec = tim) 
fluTest<- timeSeries(fluWatch[,"FluTest"], charvec = tim1) 
2. Split the dataset into two groups: training sample and test sample (as shown in Time￾Series-Forecasting-Examples.html); 
3. Follow the steps and techniques in Time-Series-Forecasting-Examples.html to answer 
the following questions. 
A. Transfer function noise modeling (10 pts) 
1) Write the mathematical equation of the fitted transfer function noise model—using 
Google Flu Trends as the explanatory variables to predict Flu Tests. 
2) Is the fitted model adequate? Explain your answer. 
Support your answers with R codes and empirical evidences. 
B. Forecast evaluation 
Consider ARIMA, TFN, NN, and NNX models. (see Time-Series-Forecasting-Examples.html 
for more details) 
1) Calculate the forecast performance in the training sample, and answer which model 
performs best in each forecast measure. 
2) Calculate the forecast performance in the test sample for the following lead time 
i. h=1 
ii. h = 4 
iii. h = 8 
iv. h = 50 
and answer which model performs best in each forecast measure. 
Support your answers with R codes and empirical evidences. 
联系我们 - QQ: 99515681 微信:codinghelp
程序辅导网!