The Multiple Linear Regression Model and Time Series Model of Language Prediction

Zhiqun Zhang¹, Changming Song², Weilian Zheng³, Zhihong Ma^4*

1College of Humanities, Tianjin Agriculture University, Tianjin, 300384, China 

2College of Engineering and Technology, Tianjin Agriculture University, Tianjin, 300384, China  

3College of Food Science and Bioengineering, Tianjin Agriculture University, Tianjin, 300384,  China 

4College of basic science, Tianjin Agricultural University, Tianjin, 300384, China

Abstract: There are currently about 6,900 languages spoken on Earth. At the same time, the total number of speakers of a language may increase ｏｒ decrease over time because of a variety of influences. We choose the quantity of the migration population, net number of tourist entry and exit, GDP per capita as three general factors which can influence the distribution of various language speakers. Then we develop a MLR model and Time series model to predict trends of global languages in the next 50 years.

Keywords: Multiple Linear Regression; Time Series Model; Language prediction