I need help with a Data Analytics question. All explanations and answers will be used to help me learn.
Enrollment forecasting is a critically important task for colleges and universities. Having a reliable forecast of future enrollment is needed to predict tuition and other revenue as well as faculty and staffing needs, course and section offerings, student services, etc.
There are many different facets to enrollment which need to be understood to best plan particular aspects of a universityâ€™s operations. For example, to properly plan course offerings, one must be able to predict the number of students in different majors and those studentsâ€™ class standing. On the other hand, revenue predictions and student services planning rely on more aggregate enrollment information.
This case study focuses on forecasting aggregate enrollment at the University for Fall 2021. The provided data set includes 8 time series. Using R Script:
- GRF2FHC = graduate student, face to face, head count (i.e. the number of students enrolled and registered in face-to-face/on campus graduate programs)
- GRF2FCR = graduate student, face to face, credits (i.e. the number of credits registered by face-to-face/on campus graduate students)
- GROLHC = graduate student, online, head count
- GROLCR = graduate student, online, credits
- UGF2FHC = undergraduate student, face to face, head count (i.e. the number of students enrolled and registered in face-to-face/on campus undergraduate programs)
- UGF2FCR = undergraduate student, face to face, credits (i.e. the number of credits registered by face-to-face/on campus undergraduate students)
- UGOLHC = undergraduate student, online, head count
- UGOLCR = undergraduate student, online, credits
Weekly data has been provided for the 19 weeks leading up to the start of the Fall 2018, 2019 and 2020 terms (APW_19, APW_18, etc.), along with the values on the first day of classes (APRSOT), the next two weeks (APW+01 and APW+02), and the â€˜census dateâ€™ (APRCEN, which is 3 weeks into the term, and is the â€˜officialâ€™ enrollment number for each semester). Some data for Fall 2021 is also available. An Excel spreadsheet with the data has been provided.
For this case you are to develop a forecast for the remaining weeks to the â€˜census dateâ€™ for Fall 2021 for each of the 8 time series. (Note that the census date prediction is the important value; however, having intermediate forecasts can help decision makers understand whether enrollment is falling below or above the prediction).
You are to present your forecast in a professional memo to the Vice President for Enrollment Management. The memo should be written in as non-technical terms as possible and should, in addition to providing the census date forecasts, also attempt to provide insight into what the forecasts mean. (For example, are there discernable trends or changes in enrollment that need to be planned for?). In addition to the memo, you should provide a technical appendix that explains, in detail, the process followed in creating your forecasts (how was the data manipulated, what models were attempted, how were they evaluated, how and why was the final model chosen). Appropriate data visualizations should be provided.