Master'sOpen Access

The comparison of quantile regression analysis and ordinary least squares methods: An application attempt

Is this your thesis?

This record came from a bulk archive import. If it’s yours, link it to your profile.

2017
0 views
0 downloads
Advisor: Prof. Dr. Hasan Durucasu

Abstract (EN)

In this thesis, factors affecting firm earnings were analyzed via quantile regression analysis and ordinary least squares in comparison. It is aimed to reveal the superiorities of the methods in modeling the business earnings which are thought to have asymmetric distribution. For this purpose, financial leverage and firm size factors which are thought to affect the business profitability were held , and effects of these two factors on different quantiles of firm earnings were examined. Various financial ratios and financial table items were used in order to measure financial leverage, firm size and firm earnings. In the study, the data of 183 firms traded on Istanbul Stock Exchange for the years 2013, 2014 and 2015 were collected, and thus sample of 536 observations were obtained. Thanks to the quantile regression analysis, the method which produces more accurate and unbiased results in the modeling of asymmetrically distributed variables, quantiles of distribution were modeled instead of modeling the conditional mean. Thus, the conclusion that firm size and financial leverage affect poor and rich firms in different directions and levels was reached. Keywords: Quantile regression, Ordinary Least Squares, Financial Leverage, Firm Size

Author

Önder Dorak

How to Cite

Önder Dorak (Master Thesis). The comparison of quantile regression analysis and ordinary least squares methods: An application attempt, 2017, Anadolu University.

License

Tüm Hakları Saklıdır

This work is shared under the specified license terms.

More theses from Anadolu University