Abstract
Extended Abstract
Introduction and Objectives: The primary objective of this research is to evaluate the efficacy of multi-factor models in predicting stock returns in the Iranian capital market. To this end, the paper aims to examine the role of diverse factors—such as firm size, book-to-market ratio, profitability, and investment policies—in explaining stock returns. By comparing these models to single-factor models, the study seeks to determine their capability in explaining return behavior within the specific conditions of the Iranian market. Accurate prediction of stock returns can provide investors, financial managers, and economic policymakers with a more precise analytical tool, thereby improving the decision-making process.
Asset pricing models have been a central focus in financial studies, with numerous efforts dedicated to enhancing their explanatory and predictive power. The first major effort in this area was the single-factor Capital Asset Pricing Model (CAPM). Despite its historical importance and widespread application, CAPM was unable to explain all variations in stock returns. This limitation led researchers to develop multi-factor models. These models incorporate additional factors beyond the market factor, such as fundamental firm characteristics and economic conditions, to increase predictive power and explanatory capability.
The capital market, as a main pillar of the financial system, faces severe fluctuations, including structural risks and susceptibility to macroeconomic developments. In this context, reliance on multi-factor models can lead to a deeper understanding of investment risks and opportunities, in addition to increasing the accuracy of stock return predictions. From an applied perspective, this is an undeniable necessity.
Methodology:This paper adopts an exploratory approach to data collection and is designed as a descriptive-correlational survey. In terms of purpose, it is classified as fundamental research. The research data is gathered based on real stock market information, including financial statements, audit reports, and data from official databases of the Tehran Stock Exchange.
The statistical population of the study consists of all companies listed on the Tehran Stock Exchange, totaling approximately 550 companies. This population was examined over a 14-year period from 2009 to 2023 (1388 to 1402 in the Persian calendar). To increase the data volume and improve the accuracy of the analysis, each fiscal year was divided into two six-month periods to track changes in stock returns and their influencing factors in greater detail.
Despite certain limitations and inconsistencies in the available company data, specific criteria were established for sample selection. Ultimately, using a systematic elimination method, a sample of 130 companies was selected for which complete and reliable data was available.
For data analysis, the data was organized and interpreted in line with the research objectives using Eviews 10 software.
Results:Based on the results of descriptive statistics tests, various financial, economic, stock market, corporate governance, and auditing variables show different patterns of stability and fluctuation. Stock returns were relatively stable with less volatility, while financial structure and liquidity levels experienced greater variability.
In the stock market domain, market value and liquidity exhibited the highest dispersion, indicating high market risk and volatility. In contrast, the market growth rate showed relative stability. Among the economic variables, inflation had the highest variability and acted as an influential variable on other indicators. Furthermore, in corporate governance, the number of board members showed more fluctuation, while ownership structure and institutional ownership demonstrated relative stability.
On the other hand, variables related to financial reporting quality and auditing had minor changes, indicating the existence of standard frameworks and procedures in this area. Due to the selection of a panel data model, the Hausman test was conducted to choose between a fixed effects model and a random effects model. The results of the Chow test for each model showed that the probability error level of this test was less than 5%, therefore confirming the fixed effects method.
In the regression analysis, the results indicate that most of the studied variables have a positive and significant relationship with stock returns. Variables such as audit process efficiency (coefficient: 0.8981), liquidity level (coefficient: 0.8978), and the market factor (coefficient: 0.8122) had a high impact, demonstrated by a t-statistic above 3 and a significance level below 0.005. Financial risk ratio and auditor expertise also had a significant effect on stock returns. The model’s coefficient of determination (R²) of 0.944 indicates its high ability to explain variations in stock returns, and the adjusted R² of 0.984 suggests the model’s acceptable predictive power.
Discussion and Conclusion:Based on the paper’s results, multi-factor asset pricing models provide an effective framework for predicting stock returns in the Iranian capital market. Unlike single-factor models like CAPM, which only consider market risk, the findings show that stock returns are influenced by a combination of financial, market, economic, governance, and auditing factors.
In the financial dimension, indicators such as capital structure, liquidity, net profit, return on assets ratio, and debt-to-asset ratio have a significant effect on returns, highlighting the importance of financial health and resource management. In the market domain, variables such as market return, growth opportunities, and book-to-market ratio, with positive and significant coefficients, confirm the role of market forces and investor expectations. From an economic perspective, factors such as the inflation rate, economic growth, and cost changes also highly influence stock performance, revealing the importance of macroeconomic conditions.
Furthermore, corporate governance structure—including board independence, ownership structure, and institutional ownership—increases transparency and reduces risk. Additionally, financial reporting quality and auditor independence are directly linked to investor confidence and improved returns.
Overall, the investigations show that generalized multi-factor models have higher explanatory power than traditional models and, by combining diverse variables, enable more accurate prediction of stock behavior. This comprehensive approach can significantly assist managers, investors, and policymakers in making informed financial decisions aligned with the stock market.
)