A Panel ARDL Application on the Research of the Relationships between Stock Returns and Financial Ratios
Cemile ÖzgürIn this research, one hundred firms that are a member of BIST Industrial Index are selected and the relationship between their stock returns and financial ratios are examined for the quarterly period of 2012-2017. For this purpose, lagged values of dependent (stock returns) and independent (financial ratios) variables are included in a regression model and the panel data is analyzed by employing Panel Autoregressive Distributed Lag (Panel-ARDL) Model that is developed during the last twenty years. The development of this method made it possible to investigate the stock return explanation and forecasting power of lagged values of all variables. According to the results obtained within this research, the existence of two different types of relationships between stock returns and financial ratios; consisting a statistically significant long-term relationship between stock returns and current ratio, accounts receivable turnover rate, asset turnover rate, net profit margin and shareholders’ equity/tangible fixed assets ratio and a statistically significant short-term relationship between stock returns and leverage ratio, asset turnover rate, net profit margin and accounts receivable turnover rate are observed.
Hisse Senedi Getirileri İle Finansal Oranlar Arasındaki İlişkinin Araştırılmasında Bir Panel ARDL Uygulaması
Cemile ÖzgürBu araştırmada, Borsa İstanbul’da hisseleri işlem gören ve BİST Sınai endeksinin birer üyesi olan yüz adet firma belirlenmiş ve bu firmaların 2012–2017 çeyrek dönemlerine ait hisse senedi getirileri ile finansal oranları arasındaki ilişki incelenmiştir. Bu amaçla, son yirmi yıl içerisinde geliştirilen ve Panel Autoregressive Distributed Lag (Panel-ARDL) olarak adlandırılan model kullanılarak, bağımlı (hisse senedi getirileri) ve bağımsız (finansal oranlar) değişkenlerin gecikmeli değerleri regresyon modeline dahil edilmiş ve oluşturulan panel veri seti analiz edilmiştir. Bu yöntemin geliştirilmesi ile tüm değişkenlerin gecikmeli değerlerinin de hisse senedi getirileri üzerindeki açıklama ve tahmin gücünün araştırılması mümkün kılınmıştır. Bu araştırma ile elde edilen bulgulara göre, hisse senedi getirileri ile cari oran, alacak devir hızı, aktif devir hızı, net kar marjı ve özsermaye/maddi duran varlıklar oranları arasında uzun dönemli; kaldıraç oranı, aktif devir hızı, net kar marjı, alacak devir hızı oranları arasında ise kısa dönemli olmak üzere, hisse senedi getirileri ile finansal oranlar arasında iki farklı türde ilişkinin mevcut olduğu görülmüştür.
Research Problem
The main aim of the study was to be able to contribute to the growing literature on the existence of the relationships between stock returns and financial ratios by employing an analysis methodology that differentiates short-term temporary relations from the long-term ones.
Research Questions
Is there a significant relationship between stock returns and lagged values of returns?
Is there a significant relationship between stock returns and their financial ratios and/or lagged values of financial ratios?
If there is; is it possible to differentiate the relationship into short-term and longterm components?
Literature Review
The purpose of the literature review was to conduct a comprehensive study in order to identify the existent research on the predictability of stock returns. Seminal works were found and classified into three main classes. The first class constituted early studies advocating the impossibility of predicting stock returns because of random walk and efficient market hypothesis. Studies mentioning about the temporary and permanent components of stock returns and advocating the predictability of returns constituted the second class. Finally, studies including the findings of predictability of returns with some macroeconomic variables and/or financial ratios were classified into the third class. Focusing on the relationships between stock returns and financial ratios, the main research findings of the literature are; there are statistically significant relationships between stock returns and financial ratios with some studies mentioning about the stock return predictive ability of financial ratios. Nevertheless, identification of a certain type of financial ratio, the extent of the relationship and a specific methodology remains open.
Methodology
In this research, a quantitative analysis methodology is employed. Stock returns and eight financial ratios of one hundred industrial firms, listed in BIST Industrial Index, are collected by using financial services software called Finnet. Using the obtained quarterly data, a panel data set is constructed for the period of 2012 – 2017. For the econometric model, the stock returns are described as a function of financial ratios and the data set is analyzed by applying an advanced method of dynamic panel estimation called Panel Autoregressive Distributed Lag (Panel-ARDL) developed by Pesaran, Shin and Smith (1999, 2001). The Model is a cointegration technique which is consistent to correct the heterogeneity bias of panel estimation and flexible enough to allow long-run coefficient homogeneity at the same time by allowing for short-run coefficient heterogeneity.
Results and Conclusions
According to the empirical results, the estimated error correction coefficient (φi ) of the developed ARDL (1,1,1,1,1,1,1,1,1) Model is negative and statistically significant which indicates the presence of a working error correction mechanism and cointegration between variables. From the estimated long-run coefficients; the Model suggests the existence of a positive and statistically significant long-term relationship between stock returns and asset turnover rate, net profit margin and shareholders’ equity/tangible fixed assets ratios, and also the existence of a negative and significant long-term relationship between stock returns, current ratio and accounts receivable turnover rate. Additionally, four statistically significant short-term variable coefficients are estimated suggesting; a positive and significant short-term relationship between stock returns and accounts receivable turnover rate, and a negative and significant shortterm relationship between stock returns and leverage ratio, asset turnover rate and net profit margin. The main implication of this research is that it shows the importance of considering the effects of periodic differences (long-term or short-term) additional to the lagged values of dependent and independent variables while analyzing the relationship between stock returns and financial ratios.