Linear and Non-linear Regression:Application to Competitor's Gasoline Volume Estimation
(2015) MASM01 20151Mathematical Statistics
- Abstract (Swedish)
- This paper is dedicated to give a better estimation for competitor's
gasoline volume on the behalf of Kalibrate Technologies and also nd
a better way to improve business and sales performance. To achieve
this goal, both linear regression model and non-linear regression model
have been used. The comparison between dierent models is discussed
in the paper. The best model provides the most accurate prediction result
based on statistical criterion, whose accuracy is greatly enhanced
compared with existing models used in Kalibrate Technologies.
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/5034934
- author
- Zhai, Haomiao
- supervisor
- organization
- course
- MASM01 20151
- year
- 2015
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- Multiple Linear Regression, Ridge Regression, LASSO, Partial Least Square Regression, Supporting Vector Regression, Random Forrest Regression
- language
- English
- id
- 5034934
- date added to LUP
- 2015-01-29 13:40:43
- date last changed
- 2015-01-29 13:40:43
@misc{5034934, abstract = {{This paper is dedicated to give a better estimation for competitor's gasoline volume on the behalf of Kalibrate Technologies and also nd a better way to improve business and sales performance. To achieve this goal, both linear regression model and non-linear regression model have been used. The comparison between dierent models is discussed in the paper. The best model provides the most accurate prediction result based on statistical criterion, whose accuracy is greatly enhanced compared with existing models used in Kalibrate Technologies.}}, author = {{Zhai, Haomiao}}, language = {{eng}}, note = {{Student Paper}}, title = {{Linear and Non-linear Regression:Application to Competitor's Gasoline Volume Estimation}}, year = {{2015}}, }