1. An individual written paper on your paper or video/webinar: (this should be about 2 pages of written text, it can be more depending on your paper and whether or not you do a demonstration of the technique)
Section #1: will briefly summarize and describe the paper’s objective.
Section #2: will briefly describe the SAS procedures/techniques (or SAS code) used, provide a description of any examples/applications and perhaps illustrate your own demo of the procedures. If you feel the method is beyond your current skill level you can just describe what they did. Some use of the SAS documentation may be required if the syntax is not well explained in the paper.
Section #3: will discuss the questions you were asked and your answers to those questions.
Section #4: will briefly describe why the paper is of interest to you and provide commentary on the paper. The commentary should include some ways the paper could be improved, and some sentences on whether you think this paper and/or technique will be useful to you or a general SAS user.
The grade will be done based on:
-Was the paper well written and grammatically correct?
-Were all sections present?
-Were written explanations clear?
-Was a demonstration done if appropriate?
For Section3 Questions asked and Answers are given are-
1) What is the objective of using the SAS Forecasting studio for E-commerce data?
· For large-scale automated forecasting, SAS Forecast Studio is a highly capable system. It has a graphical user interface for users to work with.
· It performs automatic parameter optimization and model selection, which is a time-consuming procedure in open-source and many other systems that may use the same time series models.
· Using SAS Forecast Studio reduces development time greatly.
· This system can easily manage events like public holidays or the yearly spring sale by using the front-end and back-end macros in SAS Forecast Server named PROC HPFEVENTS.
2) Can you explain the exponential smoothing technique used in this paper?
· Exponential Smoothing Methods are a family of forecasting models. They use weighted averages of past observations to forecast new values. Here, the idea is to give more importance to recent values in the series. Thus, as observations get older (in time), the importance of these values gets exponentially smaller.
· The exponential smoothing method produces a time trend forecast