Assignment 1b

.docx

School

Concordia University *

*We aren’t endorsed by this school

Course

677

Subject

Statistics

Date

Apr 3, 2024

Type

docx

Pages

2

Uploaded by DoctorComputer14490

Assignment 1b – due March 4 BSTA 677, Winter 2024 Work individually on each question and submit a single version to Moodle as a team. Use SAS (SAS programs, EG and/or Studio). You can use SAS Studio and then copy/paste the CODE into a program window of EG. This is the fastest approach since you can easier modify your code. Answer the questions below in Word and illustrate them with the results from SAS. Submit 1zipped version as a team. Questions 1) Obtain time series plot for GNP (Gross National Product) using sashelp.GNP dataset. 2) Construct a histogram and obtain basic statistics (mean, median, standard deviation, min, max) for GNP. Is this data normally distributed? 3) Does GNP series seem to have a seasonal component? Explain? 4) Which quarter of GNP is the highest, which is the lowest? Why would it be? 5) Obtain Autocorrelation plot for GNP. What does it tell you about the series? 6) Decompose GNP dataset into trend, seasonal and irregular components. Which approach: multiplicative or additive, would you recommend and why? What is the difference between these two decompositions? 7) Forecast GNP 1-year (4-step) ahead using all the following models:
a) Mean (random walk) b) Naïve (random walk) with drift c) Naïve (random walk) with drift and trend d) Naïve (random walk) with drift and trend and seasonality e) exponential smoothing models. i) Simple ii) Holt (linear) iii)Winters (additive, multiplicative) 8) Obtain 4-step ahead forecasts performance metrics such as MAD, MAPE, RSME using BACK=4 and LEAD=4 options for Holt and Winters models in 7e and select the best model. Justify your choice. Explain BACK and LEAD options.
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help