Stock Return Predictability Research Articles

Expectation disarray: Analysts' growth forecast anomaly in China

Analysts' growth forecasts positively predict stock returns in China, opposite to the results found in the US. Strategies that buy stocks with high growth forecasts and sell those with low growth forecasts earn annual abnormal returns of up to 20% (with t-values exceeding three). These results are stronger for longer-horizon forecasts and ex-ante more informative forecasts. In addition, the deviation of analysts' forecasts from unbiased forecasts based on statistical models positively predicts abnormal returns. Although the relationship between analysts' forecasts and returns in China is opposite to that in the US, these forecasts positively predict actual growth and are often too extreme, as in the US. Our results suggest that investors in China overlook valuable information contained in analysts' forecasts.

Momentum in machine learning: Evidence from the Taiwan stock market

We revisit 86 asset pricing anomalies in the Taiwan stock market and find that long-short portfolio strategies based on machine-learning methods bring substantial benefits. For example, neural networks and partial least squares generate long-short returns ranging from 1.20% to 1.50% per month. More importantly, five of the top 20 influential return predictors are momentum-related variables. This result provides novel evidence to the momentum literature given that the Taiwan stock market is one of the few exceptions to the momentum anomaly. In contrast with this conventional view, we show that momentum contributes to stock return predictability when adopting machine-learning models.

Financial indicators analysis using machine learning: Evidence from Chinese stock market

This study employs machine learning models to explore the predictive power of 10 categories of financial indicators on the Chinese stock market. We examine whether influential financial indicators fall into distinct categories of greater importance for predicting stock returns. The findings demonstrate that financial indicators across 10 categories hold predictive power for stock returns on Chinese market, with neural network models outperforming linear ones. Profitability and growth indicators are among the most influential indicators. This study contributes to a better understanding of financial indicators and demonstrates the effectiveness of machine learning models in the Chinese stock market.

ABILITY OF ACCOUNTING DATA TO PREDICT STOCK RETURNS OF NON-FINANCIAL INSTITUTIONS IN TURKEY

This study uses panel data of the top 10 non-financial companies in Turkey to analyze the impact of accounting information in predicting stock returns for period 1998 to 2019. The rationale behind choosing non-financial firms is based on the believed that financial firms have their own ways of financing and examining together with financial firms may reveal biased result. The panel least square is employed, the analysis revealed that only financial leverage, earning per share and growth opportunities have the capability of predicting stock returns in Turkey. Moreover, the outcome also revealed that a 1% increase in leverage will decline stock returns by 1.07%. On the other hand, earning per share and growth opportunity have positive effects on stock returns at 1% and 5% levels of significance, respectively. This implies that a 1% increase in earnings per share and growth opportunity will lead to an increase in stock returns by 1.8% and 2.1% respectively. However, capital expenditure and firm size are unable to predict stock returns.

Stock Return Prediction Using Voting Regressor Ensemble Learning

Abstract. The value of return on stock prices is often used in predicting profits in the process of buying and selling shares based on the calculation of the return on investment. The calculation of the value of return on stock prices can be predicted automatically at certain periods, both weekly and daily
 Purpose: The problem faced is determining a good algorithm for making predictions due to fluctuating data on stock prices making it difficult to predict.
 Methods: The stages carried out by the researcher include the data preprocessing stage and then proceed to the Exploratory Data Analysis (EDA) stage to get a pattern from the data, followed by the modeling stage on the data. This research was developed using the Python programming language where the models used to make predictions can be obtained in real-time.
 Result: The results obtained in this study show that the Voting Regressor has the best model with an error rate of 0.032523 using Root Mean Square Error (RMSE). The results of this study can be further developed to automatically predict stock return values in the future.

International stock return predictability: The role of U.S. uncertainty spillover

This paper explores the implications of U.S. uncertainty for cross-country asset pricing. We propose a global common spillover index of U.S. uncertainty (GSIU) based on the Partial Least Square method, and show that this index is a powerful predictor of aggregate stock returns around the globe, both in- and out-of-sample. Additionally, we find that GSIU affects stock returns through both cash flow and discount rate channels. Furthermore, our index is linked to well-known pricing factors, namely that an increase in GSIU is generally associated with a deteriorating macroeconomic condition, higher levels of stock market volatility and economic policy uncertainty, increased disaster risk, as well as pessimistic investor sentiment, both domestically and internationally.

ESG Rating, Investor Attention, and Stock Returns in China December 2022

We study whether ESG ratings can predict stock returns in China. We find marginal evidence that stocks with higher ESG ratings have lower future returns. In addition, we explore the cross-sectional and timeseries heterogeneities of the relationship between ESG and stock returns. We find the predictability of ESG rating is stronger for stocks in the industries that are important to climate change, and the relationship is only significant after 2018 when climate change is evident to investors in China. Overall, our findings show that investors attention is crucial for the stock return predictability of ESG ratings in China.

Inflation, interest rates and the predictability of stock returns

This paper focuses on the predictive role of inflation and interest rates in forecasting stock returns. We report economically significant in-sample evidence that stock returns are predictable using cyclical components of inflation and interest rates. The out-of-sample analysis results show that these cyclical components have reasonable out-of-sample predictability for the equity risk premium when the comparison is made against a historical average benchmark. We believe this is due to the fact that both inflation and interest rates contain stochastic trends. Accounting for changes in trend inflation (and equilibrium real interest rate) is essential for forecasting stock returns.

Stock Returns Predictability Using Financial Ratios in Malaysia

Stock return predictability is important to maintain confidence and liquidity in a stock market. The conventional theory on stock price behaviour, the Efficient Market Hypothesis, posits that stock return predictability is not possible. This study considers the Adaptive Market Hypothesis to explain the predictability of stock returns in Malaysia using financial ratios as predictor variables. The Adaptive Market Hypothesis is gaining prominence to explain stock behaviour. However, studies have been mixed and limited. This study contributes to the empirical evidence on the relationship of financial ratios to return of stock in emerging markets, specifically Malaysia. Purposive sampling is employed to select the listed companies from the Malaysia Stock Exchange, which are then filtered based on three criteria to arrive at a sample of 392 companies. For predictor variables, seven financial ratios have been identified for their predictive strength to stock returns. This study employs panel data to formulate the multiple predictive regression. The regression model determined is the fixed effects regression with robust standard error. Results show that the relationship to stock returns is significant for the predictor variables of earnings yield, dividend yield, book-to-market ratio, return on assets, current ratio and assets turnover. The relationship between debt-to-equity ratio and stock returns is not significant. The findings show that financial ratios have predictive strength to stock returns in the Malaysia market. To improve on the explanatory strength, it is suggested for future research to incorporate external factors such as GDP, inflation or other economic factors such as interest and exchange rates.

How accurate is the CAPM approach compared to the Reward Beta for shares of manufacturing and mining companies in Indonesia's LQ45 Index?

This study evaluates the efficacy of the Capital Asset Pricing Model (CAPM) and Beta Reward Model in forecasting stock returns of companies within the manufacturing and mining sectors listed on Indonesia's LQ45 index. Utilizing monthly closing stock prices from January 2010 to December 2019, the research focuses on ten companies—five from each sector—that consistently appeared in the index throughout the study period. The analysis involves a classic assumption test followed by regression analysis for each company. Key performance indicators, including R-squared, Root Mean Square Error (RMSE), and Mean Absolute Error (MAE), are employed to compare the predictive capabilities of the CAPM and Beta Reward Model. The findings indicate a systematic and superior performance of the Beta Reward Model over the CAPM in predicting stock returns in the Indonesian context. This study contributes to the existing literature on stock return prediction models and provides practical insights for investors and financial analysts in Indonesia.

Adaptive Market Hypothesis and Predictability: Evidence in Latin American Stock Indices

AbstractThis article examines the adaptive market hypothesis in the five most important Latin American stock indices. To that end, we apply three versions of the variance ratio test, as well as the Brock-Dechert-Scheinkman test for nonlinear predictability. Additionally, we perform the Dominguez-Lobato and generalized spectral tests to evaluate the Martingale difference hypothesis. Moreover, we consider salient news related to the plausible market inefficiencies detected by these four tests. Finally, we apply a GARCH-M model to assess the risk-return relationship through time. Our results suggest that the predictability of stock returns varies over time. Furthermore, the efficiency in each market behaves differently over time. All in all, the analyzed emerging market indices satisfy the adaptive market hypothesis, given the switching behavior between periods of efficiencies and inefficiencies, since the adaptive market hypothesis suggests that market efficiency and market anomalies might coexist in capital markets.

High-frequency stock return prediction using state-of-the-art deep learning models

Determining stock price movements is a challenging problem because stock prices are often influenced by multiple factors such as economic, political, business, and human behavior. In this paper, we will attempt different modeling methods for two types of data, a total of 40 Dow Jones Industrial Index components, to verify the effectiveness of daily and high-frequency data for stock price prediction. Furthermore, we will attempt to validate the performance of LSTM model in stock price prediction, and also try to improve its performance by incorporating an attention mechanism. We assume that adding an attention layer to LSTM model would improve model performance in our data sets, especially in high-frequency data, since the data set would contain a huge amount of noise. Our results indicate that the simple LSTM performs better than the attention-based LSTM for both data types of prediction tasks with a benchmark of the number of stock prediction outcomes that outperform the number of those in other model, which is 24 out 40 stocks, which refutes our initial assumptions and does not validate whether adding attention mechanism is useful for solving the shallow layers and gradient vanishing problem and thus improving the LSTM model performance.

Relative informative power and stock return predictability: a new perspective from Egypt

Purpose This paper aims to introduce a new perspective on long-term stock return predictability by focusing on the relative (individual and hybrid) informative power of a wide range of accounting (firm-related), technical and macroeconomic factors while considering the past performance of the stocks using machine learning algorithms. Design/methodology/approach The sample includes a panel data set of 94 non-financial firms listed in Egyptian Exchange 100 index from 2014: Q1 to 2019: Q4. Relativity has been investigated by comparing relevant factors’ individual and combined informative power and differentiating between losers and winners based on historical stock returns. To predict the quarterly stock returns, Gaussian process regression (GPR) has been used. The robustness of the results is examined through the out-of-sample test. This study also uses linear regression (LR) as a benchmark model. Findings The past performance and the presence of other predictors influence the informative power of relevant factors and hence their predictive ability. The out-of-sample results show a trade-off between GPR and LR with proven superiority to GPR in limited experiments. The individual informative power outperforms the hybrid power, in which macroeconomic indicators outperform the remaining sets of indicators for losers, while winners show mixed results in terms of various performance evaluation metrics. Prediction accuracy is generally higher for losers than for winners. Practical implications This study provides interesting insight into the dynamic nature of the predictor variables in terms of stock return predictability. Hence, this study also deepens the understanding of asset pricing in a way that directly contributes to practitioners’ portfolio diversification strategies. Originality/value In concern of the chaos of factors in the literature and its accompanying misleading conclusions, this study takes another look at the approach that studies stock return predictability. To the best of the authors’ knowledge, this is the first study in the Egyptian context that re-examines the predictive power of the previously discovered factors from a different perspective that highlights their relative nature.

Leveraging Return Prediction Approaches for Improved Value-at-Risk Estimation

Value at risk is a statistic used to anticipate the largest possible losses over a specific time frame and within some level of confidence, usually 95% or 99%. For risk management and regulators, it offers a solution for trustworthy quantitative risk management tools. VaR has become the most widely used and accepted indicator of downside risk. Today, commercial banks and financial institutions utilize it as a tool to estimate the size and probability of upcoming losses in portfolios and, as a result, to estimate and manage the degree of risk exposure. The goal is to obtain the average number of VaR “failures” or “breaches” (losses that are more than the VaR) as near to the target rate as possible. It is also desired that the losses be evenly distributed as possible. VaR can be modeled in a variety of ways. The simplest method is to estimate volatility based on prior returns according to the assumption that volatility is constant. Otherwise, the volatility process can be modeled using the GARCH model. Machine learning techniques have been used in recent years to carry out stock market forecasts based on historical time series. A machine learning system is often trained on an in-sample dataset, where it can adjust and improve specific hyperparameters in accordance with the underlying metric. The trained model is tested on an out-of-sample dataset. We compared the baselines for the VaR estimation of a day (d) according to different metrics (i) to their respective variants that included stock return forecast information of d and stock return data of the days before d and (ii) to a GARCH model that included return prediction information of d and stock return data of the days before d. Various strategies such as ARIMA and a proposed ensemble of regressors have been employed to predict stock returns. We observed that the versions of the univariate techniques and GARCH integrated with return predictions outperformed the baselines in four different marketplaces.

Predicting stock market returns using aggregate credit risk

We investigate how credit risk predicts stock returns in the time-series at the aggregate level in the Chinese market. We find that the aggregate credit risk, measured by the option-based structural model, is a strong positive predictor of future stock market excess returns at various horizons. The predictive power remains significant even after controlling for a number of widely-researched predictors or under out-of-sample tests. The positive relationship between aggregate credit risk and expected stock market returns accords with the risk-return tradeoff theory. We also find that the predictive power comes from the discount rate channel. A higher level of aggregate credit risk is related to a higher discount rate of future cash flows, and thus generates higher expected returns.

Moments, shocks and spillovers in Markov-switching VAR models

To investigate how economies, financial markets or institutions can deal with stress, we often analyze the effects of shocks conditional on being in a recession or a bear market. MSVAR models are perfectly suited for such analyses because they combine gradual movements with sudden regime switches. In this paper, we develop a comprehensive methodology to conduct these analyses. We derive first and second moments conditional only on the regime distribution and propose impulse response functions for both moments. By formulating the MSVAR as an extended linear non-Gaussian VAR, all results are available in closed-form. We illustrate our methods with an application to stock and bond return predictability. We show how forecasts of means, volatilities and (auto-)correlations depend on the regimes. The effect of shocks becomes highly nonlinear, and they propagate via different channels. During bear markets, shocks have stronger effects on means and volatilities and die out more slowly.

Determinan return saham pada perusahaan manufaktur periode 2016-2020 dengan analisis rasio keuangan

This study aims to analyze the effect of liquidity, profitability, and solvency on stock returns in manufacturing companies for the 2016-2020 period. The data in this study uses secondary data in the form of financial reports that meet the criteria. The technique used in the analysis in this study was panel data linear regression using the EVIEWS 10 program. The test results using panel data regression with the EVIEWS 10 tool yielded only two of the three independent variables that had a significant effect on stock returns. Profitability as measured by return on equity and solvency as measured by the debt to equity ratio predicts stock returns in a positive direction. Meanwhile, liquidity as measured by the current ratio has no effect on stock returns. Investment in the manufacturing sector which is a capital-intensive company requires financing to develop the business so that a high solvency value can reflect a better stock return.

ANALYSIS OF SHARE PERFORMANCE USING THREE FACTORS MODEL FAMA AND FRENCH (TFMFF) IN ILQ-45 STOCK PERIOD 2017-2020

Research conducted by Fama and French in 1996 showed that there were factors other than the beta that were significantly able to predict stock returns. However, several subsequent studies showed inconsistent results. The discrepancy between the results of previous studies prompted this research to be carried out. In this study, the researchers selected the stocks used based on the criteria for company profits and return on equity (ROE) offered by Warren Buffett. This study uses Buffett's criteria in selecting stocks that are used to analize TFMFF in estimating the return of stocks grouped into the Fama French portfolio. The method used is quantitative. The secondary data used are quarterly close price data, mining company equity value, Bank Indonesia interest rate (risk-free rate), and number of outstanding shares (number of shares outstanding). The results show that the TFMFF is accurate in predicting stock returns.

Improved tests for stock return predictability

– Predictive regression methods are widely used to examine the predictability of (excess) stock returns by lagged financial variables characterized by unknown degrees of persistence and endogeneity. We develop a new hybrid test for predictability in these circumstances based on simple regression t-statistics. Where the predictor is endogenous, the optimal, but infeasible, test for predictability is based on the t-statistic on the lagged predictor in the basic predictive regression augmented with the current period innovation driving the predictor. We propose a feasible version of this augmented test, designed for the case where the predictor is an endogenous near-unit root process, using a GLS-based estimate of the innovation used in the infeasible test regression. The limiting null distribution of this statistic depends on both the endogeneity correlation parameter and the local-to-unity parameter characterizing the predictor. A method for obtaining asymptotic critical values is discussed and response surfaces are provided. We compare the asymptotic power properties of the feasible augmented test with those of a (non augmented) t-test recently considered in Harvey et al. and show that the augmented test is more powerful in the strongly persistent predictor case. We then propose using a weighted combination of the augmented statistic and the t-statistic of Harvey et al., where the weights are obtained using the p-values from a unit root test on the predictor. We find this can further improve asymptotic power in cases where the predictor has persistence at or close to that of a unit root process. Our final hybrid testing procedure then embeds the weighted statistic within a switching-based procedure which makes use of a standard predictive regression t-test, compared with standard normal critical values, when there is evidence for the predictor being weakly persistent. Monte Carlo simulations suggest that overall our new hybrid test displays superior finite sample performance to comparable extant tests.

Which is more important in stock market forecasting: Attention or sentiment?

We examine the predictability of investors' attention and sentiment on four major stock indexes' returns in China between 2008 and 2018. We create investors' attention and sentiment variables by using the investors' posts from Guba, which is the largest stock forum in China. We find that both investor attention and sentiment can improve the stock return predictability with in-sample and out-of-sample periods. However, investor sentiment is more informative than investor attention in stock return prediction. These results remain significant by testing posts announced in trading and non-trading sessions. Our paper provides new empirical evidence for stock return forecasting in a data-rich environment.