چکیده:
In financial markets such as Tehran Stock Exchange, P/E coefficient, which is one
of the most well-known instruments for evaluating stock prices in financial markets,
is considered necessary for shareholders, investors, analysts and corporate
executives. P/E is used as an important indicator in investment decisions. In this
research, harmony search metaheuristic algorithm is used to select optimal variables
affecting P/E and then, modelling is done through multivariate regression
based on panel data. For this purpose, a sample of 87 companies has been selected
from listed companies in the Tehran Stock Exchange during a 10-year period
(2006-2015). The results indicate the effect of the variables of stock returns, stock
price to book value ratio, price to net selling ratio, return on assets, earnings per
share, market value to book value, money volume, operating return margin, return
on capital, and current assets, as top ten variables, on P/E ratio, which estimates a
total of 86% of the P/E ratio changes.
خلاصه ماشینی:
ARTICLE INFO Article history: Received 22 April 2017 Accepted 12 January 2018 Keywords: Metaheuristic algorithms, Harmony search, Forecasting price to earnings (P/E) ratio, Econometric of panel data ABSTRACT In financial markets such as Tehran Stock Exchange, P/E coefficient, which is one of the most well-known instruments for evaluating stock prices in financial mar- kets, is considered necessary for shareholders, investors, analysts and corporate executives.
Considering that forecast accuracy is the most important factor in choosing any forecast method, Sureshkumar and Elango [19] used forecast algorithms and functions to forecast the stock price in the future of the Indian market and compared performance of these methods.
2. 4 Research Questions The main question of this research is how to use meta-intuitive patterns such as harmony search in explaining and presenting models of forecasting financial and economic variables such as P/E in the econometric approach of panel data.
In order to answer the research question, the following hypothe- sis is presented: The first hypothesis: there is the possibility of designing and explaining the P/E price forecast model using harmony search algorithm in optimizing the panel-based econometric model 3 Metaheuristic Algorithms The process of designing and implementing metaheuristic algorithms has three successive stages, each of which has different steps.
In order to better under- stand the behaviour of stock prices in future research, a combination of fundamental and technical analysis, as well as intelligent algorithms such as Cuckoo algorithm, Bird Algorithm and the growth of the bacteria are suggested to be used to forecast the P/E ratio and compare their performance with harmony search neural network.