GPU Implementation of Sales Forecasting with Linear Regression

Ayomide Yusuf; Shadi Alawneh

doi:10.21276/ijircst.2018.6.4.1

Abstract

Forecasting of sales is very important in any business as it helps managers to learn from historical data and make informed decisions. This generally involves intensive processes using spreadsheets which require inputs from all levels in an organization. This approach introduces bias and is generally not accurate. There are several methods that have been used in the past to forecast sales, such as Exponential Smoothing, Moving Average, and Autoregressive Moving Average (ARMA). Due to the nature of the data, it usually takes more time for these methods to analyze the sales data and make predictions. In this paper, the sales data is analyzed and predictions are made by using linear regression as implemented on the GPU to make the process faster. Sales forecasting is made possible by finding best fit line by linear regression techniques (i.e. linear convolution). To illustrate this process, simulated sales data was used. The sales forecasting with linear regression implementation using GPU was compared to the CPU implementation and a speedup of up to 7.557x was achieved.

Keywords

GPU, CUDA, Linear Regression, Sales Forecasting.

References

[1] http://hbr.org/1971/07/how-to-choose-the-right-forecas ting-technique

[2] Sales forecasting is not about complex algorithm.

[3] http://www.yourarticlelibrary.com/sales/sales-forecasti ng-top-9-methods-of-sales-forecasting/50998

[4] https://blog.getbase.com/5-essential-sales-forecasting-t echniques

[5] https://en.wikipedia.org/wiki/Linear_regression

[6] Li Bing-jun, He Chun-hua, China, “The Combined Forecasting Method of GM(1,1) with Linear Regression and Its Application” 2007 IEEE International Conference.

[7] “Programming with CUDA”, www.nvidia.com.

[8] Jyoti B. Kulkarni1, A. A. Sawant2, Vandana S. Inamdar3 Database Processing by Linear Regression on GPU using CUDA

[9] “Getting Started with CUDA”, www.nvidia.com

[10] Shadi Alawneh and Dennis Peters, Proc. 14th IEEE International Conference on High Performance Computing and Communications (HPCC-2012), June 2012, Liverpool, UK.

[11] https://en.wikipedia.org/wiki/CUDA

[12]SamanehBeheshti-Kashi, Hamid Reza Karimi,Klaus-Dieter Thoben, Michael Lütjen and Michael Teucke, “A survey on retail sales forecasting and prediction in fashion markets, ” Systems Science & Control Engineering, Volume 3, 2015.

[13] Henrik Aronsson and Rickard Jonsson, “Sales forecasting Management,” A bachelor thesis in Management Accounting, 2008.

[14] Rashmi Sharma and Ashok K. Sinha, “Sales forecasting of an automobile industry,” International Journal of Computer Applications

[15] http://docs.nvidia.com/cuda/cusparse/index.html

[16] http://docs.nvidia.com/cuda/curand/index.html

[17] http://docs.nvidia.com/cuda/thrust/index.html

[18]http://docs.nvidia.com/cuda/npp/index.html

[19]http://openacc.org

[20]David B. Kirk and Wen-mei W. Hwu, Programming Massively Parallel Processor

Cites this article as

A. Yusuf, S. Alawneh, "GPU Implementation of Sales Forecasting with Linear Regression", International Journal of Innovative Research in Engineering & Management (ijircst), Vol-6, Issue-4, Page No-43-48, 2018. Available from: https://doi.org/10.21276/ijircst.2018.6.4.1

Corresponding Author

Ayomide Yusuf

Electrical and Computer Engineering, Oakland University, Rochester, MI, USA. E-mail: ayomideyusuf @oakland.edu

Download Full Paper

Download PDF

No. of Downloads: 20 | No. of Views: 2290

GPU Implementation of Sales Forecasting with Linear Regression

Citations

Download Full Paper PDF

Total View 2290

Total Download 20