Transcript Group D

FORECASTING FOR
GLINA ARTS
Asha Gangineni
Samyukta Varre
Gowtham Laburam
Rahul Sawant
Problem statement
■ Prediction of Total enrollment of students for Fall -2016
■ Prediction of Enrollment of students by media for Fall-2016
■ Prediction of enrollment of Students by Courses for Fall-2016
Assumptions
■ We incremented the Data by one year, EG: Fall 2014 became Fall 2015
■ Data is available till Spring 2015 , so we incremented the data by one year such that
the prediction can be made for fall 2016
■ No Specified Season was changed to Fall 2014
■ Registration count <5 is not considered
■ 2D and No media Specified were not considered as data was not available for 5
seasons
Approach and Technology Considered
Single Exponential Smoothing:
■ Suitable when there is no trend or
seasonal pattern
■ Forecast for a small period of time.
Selecting the Target Element
Tr
Trend Line For Months
■
R^2 determines the strength of relationship( values are between 0 and 1)
■
Closer to 1 then there is a good relationship between the factors that we are comparing.
Trend Line For Quarters
Statistical Formulae Used
Mean Absolute Error:
■ The MAE measures the average magnitude of the errors in a set of forecast.
Mean Absolute Percentage Error:
■ Mean Absolute Percent Error (MAPE) is the most common measure of forecast error.
■ The MAPE (Mean Absolute Percent Error) measures the size of the error in
percentage terms
Mean Square Error:
■ The Mean Squared Error (MSE) is a measure of how close a fitted line is to data
points
Statistical Formulae Used
■ Forecasting Exponential Smoothing
Ft+1 = α At +(1- α) Ft
where
α = Smoothing Factor
At = Last Actual value
Ft = Last Forecast Value
Exponential Smoothing by seasonAdults
Forecast for clay for Fall-2016
CLAY
Throwing on the wheel
Clay
Clay Extended Access
Clay Open Studio
Forecast for Fiber for Fall-2016
Fiber
Craft Uncorked - Dyed
Silk Scarf
Forecast for Glass for Fall-2016
Glass
Craft Uncorked Fused Coster Glass
Forecast for Graphics for Fall-2016
Graphics
Emerson Family Saturday
Forecast for Metals for Fall-2016
Metals
Jewelry Foundation
Forecast for Metals for Fall-2016
Craft Uncorked Wired
Forecast for Wood Turning for Fall-2016
Wood Turning
Camps
Forecasting for Camps for Fall-2016
Forecasting for Camps- Fall 2016
Camps
Metal Smith for kids
Forecast for Child and Parent for Fall2016
Forecast for Child Camp for Fall-2016
Child camp
Learning the Potter’s wheel
Forecast for Clay for Fall-2016
Clay
Throwing on the wheel
Forecast for Wood Turning for Fall-2016
Wood turning
Beginning wood turning
Conclusion - Registration by Media
Conclusion - Registration by Course
Recommendations
■ No sufficient data , Missing links in data
■ Better forecasting could be done id data was available for more than 4 years
■ Maintain two different columns for Registration and enrollment dates for more
accurate prediction
■ Provide recent data
Questions ?