Variable selection is one of the most important steps in statistical modeling.For this purpose, typically backward elimination technique is used. Since in these methods two stage model fitting and variable selection done separately, the results is unstable. Recently, another group of variable selection methods called shrinkage methods have been proposed. Among them, LASSO Enjoys special popularity. In this article, the most important fundamental accounting variables affecting on stock return of listed companies in Tehran Stock Exchange identified by LASSO in R in the period 1386-1390.Based on the findings, the return on assets is selected in all the years. In contrast three variables investment, gross profit margin and changes in equity were not selected in any years. The Simplest and most complex models belong to 1388 and 1389, respectively with two and six variables.
Seyednezhad Fahim,S. R. , Sohrabi,N. and Movaghari,H. (2015). Stock Return Prediction via LASSO Shrinkage Method. Accounting and Auditing Studies, 4(13), 40-53. doi: 10.22034/iaas.2015.103472
MLA
Seyednezhad Fahim,S. R. , , Sohrabi,N. , and Movaghari,H. . "Stock Return Prediction via LASSO Shrinkage Method", Accounting and Auditing Studies, 4, 13, 2015, 40-53. doi: 10.22034/iaas.2015.103472
HARVARD
Seyednezhad Fahim S. R., Sohrabi N., Movaghari H. (2015). 'Stock Return Prediction via LASSO Shrinkage Method', Accounting and Auditing Studies, 4(13), pp. 40-53. doi: 10.22034/iaas.2015.103472
CHICAGO
S. R. Seyednezhad Fahim, N. Sohrabi and H. Movaghari, "Stock Return Prediction via LASSO Shrinkage Method," Accounting and Auditing Studies, 4 13 (2015): 40-53, doi: 10.22034/iaas.2015.103472
VANCOUVER
Seyednezhad Fahim S. R., Sohrabi N., Movaghari H. Stock Return Prediction via LASSO Shrinkage Method. Accounting and Auditing Studies, 2015; 4(13): 40-53. doi: 10.22034/iaas.2015.103472
