ﻛﺎرﺑﺮد روش ﺗﺮﻛﻴﺒﻲ ﻣﺎﺷﻴﻦ ﺑﺮدار ﭘﺸﺘﻴﺒﺎن و اﻧﺘﺨﺎب وﻳﮋﮔﻲ ﺑﻪ ﻣﻨﻈﻮر ﭘﻴﺶﺑﻴﻨﻲ درﻣﺎﻧﺪﮔﻲ ﻣﺎﻟﻲ ﺷﺮﻛﺖ ﻫﺎي ﭘﺬﻳﺮﻓﺘﻪ ﺷﺪه در ﺑﻮرس اوراق ﺑﻬﺎدار ﺗﻬﺮان

Use of Combined Approach of Support Vector Machine and Feature Selection for Financial Distress Prediction of Listed Companies in Tehran Stock Exchange Market

ﭘﻴﺶﺑﻴﻨﻲ درﻣﺎﻧﺪﮔﻲ ﻣﺎﻟﻲ از ﻣﺴﺎﺋﻞ ﻣﻬﻤﻲ اﺳﺖ ﻛﻪ ﻫﻤﻮاره ﭘﮋوﻫﺸﮕﺮان، ﻣﺆﺳﺴﻪ ﻫـﺎي اﻋﺘﺒﺎري و ﺑﺎﻧﻚﻫﺎ ﺑﻪ آن ﺗﻮﺟﻪ ﻛﺮده اﻧﺪ. ﺗﺎﻛﻨﻮن ﺗﺤﻘﻴﻘﺎت ﺑﺴﻴﺎري در اﻳﻦ زﻣﻴﻨـﻪ ﺻـﻮرت ﮔﺮﻓﺘـﻪ اﺳﺖ، وﻟﻲ اﺳﺘﻔﺎده از ﻣﺪلﻫﺎي ﺗﺮﻛﻴﺐ ﺷﺪة اﻧﺘﺨﺎب وﻳﮋﮔﻲ و ﻣﺪل ﻃﺒﻘﻪ ﺑﻨـﺪي ﻛﻨﻨـﺪه ، از ﻣﺴـﺎﺋﻠﻲ اﺳﺖ ﻛﻪ ﻓﻘﻂ در ﺳﺎلﻫﺎي اﺧﻴﺮ ﺗﻮﺟﻪ ﭘﮋوﻫﺸﮕﺮان را ﺑﻪ ﺧﻮد ﺟﻠـﺐ ﻛـﺮده اﺳـ ﺖ. در اﻳـﻦ ﻣﻘﺎﻟـﻪ ﻣﺎﺷﻴﻦ ﺑﺮدار ﭘﺸﺘﻴﺒﺎن ﺑﺎ ﭼﻬﺎر ﺗﺎﺑﻊ ﻛﺮﻧﻞ ﺧﻄﻲ، ﭼﻨﺪﺟﻤﻠﻪاي، ﺷﻌﺎﻋﻲ و ﺳﻴﮕﻤﻮﻳﻴﺪ ﺑﻪ ﻋﻨﻮال ﻣـﺪل ﻃﺒﻘﻪ ﺑﻨﺪيﻛﻨﻨﺪه و ﺗﺮﻛﻴﺐ آن ﺑﺎ روشﻫﺎي اﻧﺘﺨﺎب وﻳﮋﮔﻲ ﻓﻴﻠﺘﺮﻛﻨﻨﺪه و ﭘﻮﺷـﺶ دﻫﻨـﺪه اﺳـﺘﻔﺎده ﺷﺪه اﺳﺖ. ﻫﻤﭽﻨﻴﻦ از اﻟﮕﻮرﻳﺘﻢ ژﻧﺘﻴﻚ ﻛﻪ ﻳﻜﻲ از اﻧﻮاع روشﻫﺎي ﭘﻮﺷﺶدﻫﻨﺪة اﻧﺘﺨﺎب وﻳﮋﮔﻲ اﺳﺖ و روشﻫﺎي آﻧﺎﻟﻴﺰ اﺟﺰاي اﺳﺎﺳﻲ، زﻧﺠﻴﺮة اﻃﻼﻋﺎت و رﻟﻴﻒ ﻛﻪ ﺟﺰء روشﻫـﺎي ﻓﻴﻠﺘﺮﻛﻨﻨـﺪ ة اﻧﺘﺨﺎب وﻳﮋﮔﻲ ﻫﺴﺘﻨﺪ، اﺳﺘﻔﺎده ﺷﺪه اﺳﺖ. ﻧﺘﺎﻳﺞ ﺑﻪ دﺳـﺖ آﻣـﺪه ﻧﺸـﺎن داد ﻛـﻪ روش اﻟﮕـﻮرﻳﺘﻢ ژﻧﺘﻴﻚ ﻧﺴﺒﺖ ﺑﻪ روشﻫﺎي ﻓﻴﻠﺘﺮﻛﻨﻨـﺪه ، ﻋﻤﻠﻜـﺮد ﺑﻬﺘـﺮي دارد . ﻫﻤﭽﻨـﻴﻦ دﻗـﺖ ﻣﺎﺷـﻴﻦ ﺑـﺮدا ر ﭘﺸﺘﻴﺒﺎن ﺑﺎ ﺗﻮاﺑﻊ ﻛﺮﻧﻞ ﺧﻄﻲ، ﭼﻨﺪﺟﻤﻠﻪاي، ﺷﻌﺎﻋﻲ و ﺳﻴﮕﻤﻮﻳﻴﺪ در ﺗﺮﻛﻴﺐ ﺑﺎ اﻟﮕﻮرﻳﺘﻢ ژﻧﺘﻴﻚ، ﺑـﺎ ﺳﻄﺢ اﻃﻤﻴﻨﺎن 95 درﺻﺪ ﺗﻔﺎوت ﻣﻌﻨﺎداري ﺑﺎ ﻫﻢ ﻧﺪارﻧﺪ.

Financial distress prediction (FDP) is a great important subject that has always been interesting to researchers, financial institutions and banks. Tough many works have been done in this area, but use of combined approach of feature selection and classifier is an issue that has attracted researchers' attention just in recent years. In this paper, four well-known kinds of SVM that each of them has it's own kernel function including: linear, polynomial, radial and sigmoid have been introduced as the main classifiers of our proposed approach. These four methods have been integrated with genetic algorithm (GA) as a wrapper feature selection approach as well as three techniques of filtering feature selection approach called: principle component analysis (PCA), information gain and relief. Brought results indicated that genetic algorithm outperformed the other feature selection techniques in it's combination with SVM methods. Furthermore, implemented hypothesis test implied that there was no significance level among GA-SVM (linear), GA-SVM (radial), GA-SVM (polynomial) and GA-SVM (sigmoid) techniques with confidence level of %95.