Conditional Factor Models Research Articles

Recurrent double-conditional factor model

AbstractIn economic applications, the behavior of objects (e.g., individuals, firms, or households) is often modeled as a function of microeconomic and/or macroeconomic conditions. While macroeconomic conditions are common to all objects and change only over time, microeconomic conditions are object-specific and thus vary both among objects and through time. The simultaneous modeling of microeconomic and macroeconomic conditions has proven to be extremely difficult for these applications due to the mismatch of dimensions, potential interactions, and the high number of parameters to estimate. By marrying recurrent neural networks with conditional factor models, we propose a new white-box machine learning method, the recurrent double-conditional factor model (RDCFM), which allows for the modeling of the simultaneous and combined influence of micro- and macroeconomic conditions while being parsimoniously parameterized. Due to the low degree of parameterization, the RDCFM generalizes well and estimation remains feasible even if the time-series and the cross-section are large. We demonstrate the suitability of our method using an application from the financial economics literature.

Adaptive Testing for Alphas in Conditional Factor Models with High Dimensional Assets

This article focuses on testing for the presence of alpha in time-varying factor pricing models, specifically when the number of securities N is larger than the time dimension of the return series T. We introduce a maximum-type test that performs well in scenarios where the alternative hypothesis is sparse. We establish the limit null distribution of the proposed maximum-type test statistic and demonstrate its asymptotic independence from the sum-type test statistics proposed by Ma et al. Additionally, we propose an adaptive test by combining the maximum-type test and sum-type test, and we show its advantages under various alternative hypotheses through simulation studies and two real data applications.

On the conditional performance of the IVOL anomaly

This paper proposes that the well-known idiosyncratic volatility (IVOL) puzzle is partly due to an overconditioning bias in IVOL estimates when there exists nonlinearity between stock and factor returns. We analytically derive the overconditioning bias when beta and IVOL are contemporaneously estimated using daily returns. To mitigate the estimation bias, we employ a conditional factor model, as in Avramov and Chordia (2006), to estimate systematic risk exposure and IVOL. Our empirical results show that the conditional IVOL estimates do not imply a negative return premium when firm fundamentals are used as conditioning variables, and the associated conditional factor-adjusted alpha is much smaller than the original IVOL puzzle. Our findings are robust in the U.S. and Chinese stock markets.

Macroeconomic factor, firm characteristics and inventory holding in Nigeria: A quantile regression approach

Purpose: Prior studies show that inventory holding is closely linked to liquidity and procyclical dependent on the combination of macroeconomic and firm characteristics. Thus, conditional linear factor models such as OLS should fail to explain the inventory-holding motive, especially in the context of developing countries. This study seeks to empirically investigate the determinants of corporate inventory holding based on evidence from pharmaceutical companies in Nigeria. Research methodology: The study adopts the ex post facto research design. The final sample was eight pharmaceutical & healthcare firms quoted on the Nigerian Stock Exchange (NSE). The data were analysed using the quantile regression technique. Results: The results showed that the inflation rate had a positive effect on the inventory holding distribution at upper quantiles (75th); and, the cash conversion cycle on the inventory holding was significant at different quantiles (25th, 50th and 75th). Profitability and liquidity were non-significant at different quantile distributions. Limitations: The focus on pharmaceutical firms limits the generalizability of the study findings to other sectors of the economy. Contributions: The study contributes to the literature in the context of developing countries, on the impact of varying firm characteristics and inflation rates on the different conditional distribution of the regressand, i.e., inventory holding.

Modeling Corporate Bond Returns

ABSTRACTWe propose a conditional factor model for corporate bond returns with five factors and time‐varying factor loadings. We have three main empirical findings. First, our factor model excels in describing the risks and returns of corporate bonds, improving over previously proposed models in the literature by a large margin. Second, our model recommends a systematic bond investment portfolio whose high out‐of‐sample Sharpe ratio suggests that the credit risk premium is notably larger than previously estimated. Third, we find closer integration between debt and equity markets than found in prior literature.

Systematic Contagion Effects of the Global Finance Crisis: Evidence from the World’s Largest Advanced and Emerging Equity Markets

This paper examines the systematic contagion effects of the global financial crisis of 2007–2009 on the world’s largest advanced and emerging equity markets, using the conditional factor model of Dungey and Renault (2018) and and the adjusted correlation coefficient approach of Forbes and Rigobon (2002). Our findings indicate that when applying the Forbes and Rigobon approach, no evidence of contagion is found, while using the conditional factor model, we observe significant evidence of contagion in the aggregate equity markets of both advanced and emerging markets. Furthermore, the results from the conditional factor model suggest that the structural relationship across the financial sectors of advanced and emerging markets was significantly disrupted during the crisis period.

Robust high-dimensional alpha test for conditional time-varying factor models

This paper considers the problem of testing the presence of alpha in high-dimensional conditional time-varying factor model. We proposed a spatial-sign-based test procedure which is robust and efficient for heavy-tailed distributions. We established the theoretical properties of the proposed test statistic under some mild conditions. Simulation studies and a real data example also show the superior of our test procedure to the existing methods.

Robust Estimation of Conditional Factor Models

This paper develops estimation and inference methods for conditional quantile factor models. We first introduce a simple sieve estimation, and establish asymptotic properties of the estimators under large $N$. We then provide a bootstrap procedure for estimating the distributions of the estimators. We also provide two consistent estimators for the number of factors. The methods allow us not only to estimate conditional factor structures of distributions of asset returns utilizing characteristics, but also to conduct robust inference in conditional factor models, which enables us to analyze the cross section of asset returns with heavy tails. We apply the methods to analyze the cross section of individual US stock returns.

A Unified Framework for Estimation of High-dimensional Conditional Factor Models

A Unified Framework for Estimation of High-dimensional Conditional Factor Models

Conditional Time-Varying General Dynamic Factor Models and Its Application to the Measurement of Volatility Spillovers across Russian Assets

This study analyzes the spillover effects of volatility in the Russian stock market. The paper applies the Diebold–Yilmaz connectedness methodology to characterize volatility spillovers between Russian assets. The spectral representation of the forecast variance decomposition proposed by Baruník and Křehlik is used to describe the connectivity in short-term (up to 5 days), medium-term (6–20 days) and long-term (more than 20 days) time frequencies. Additionally, two new augmented models are developed and applied to evaluate conditional spillover effects in different sectors of the Russian economy for the period from January 2012 to June 2021. It is shown that spillover effects increase significantly during political and economic crises and decrease during periods of relative stability. The rising of the overall level of spillovers in the Russian stock market coincides in time with the political crisis of 2014, the intensification of anti-Russian sanctions in 2018 and the fall in oil prices and the start of the pandemic in 2020. With the consideration of the augmented models it can be argued that a significant part of the long-term spillover effects on the Russian stock market may be caused by the influence of external economic and political factors. However, volatility spillovers generated by internal Russian idiosyncratic shocks are short-term. Thus, the proposed approach provides new information on the impact of external factors on volatility spillovers in the Russian stock market.

Firm Characteristics and Global Stock Returns: A Conditional Asset Pricing Model

Abstract This paper studies the relation between 36 firm-level characteristics and stock returns in 48 countries using instrumented principal components analysis. A non-U.S. country-neutral conditional factor model performs well in describing risk and returns and generates small and statistically insignificant anomaly intercepts when allowing for three or more latent factors. The non-U.S. model performs better in emerging than in developed markets, while showing substantial differences across countries. On average, only 10 characteristics significantly contribute to the models’ performance. Market beta, momentum, and firm size characteristics instrument for systemic exposure in U.S. and non-U.S. models, while investment and book-to-market do not. (JEL G11, G12, G14, G15) Received January 28, 2021; editorial decision July 30, 2021

Flexible Time-Varying Betas in a Novel Mixture Innovation Factor Model with Latent Threshold

This paper introduces a new methodology to estimate time-varying alphas and betas in conditional factor models, which allows substantial flexibility in a time-varying framework. To circumvent problems associated with the previous approaches, we introduce a Bayesian time-varying parameter model where innovations of the state equation have a spike-and-slab mixture distribution. The mixture distribution specifies two states with a specific probability. In the first state, the innovation variance is set close to zero with a certain probability and parameters stay relatively constant. In the second state, the innovation variance is large and the change in parameters is normally distributed with mean zero and a given variance. The latent state is specified with a threshold that governs the state change. We allow a separate threshold for each parameter; thus, the parameters may shift in an unsynchronized manner such that the model moves from one state to another when the change in the parameter exceeds the threshold and vice versa. This approach offers great flexibility and nests a plethora of other time-varying model specifications, allowing us to assess whether the betas of conditional factor models evolve gradually over time or display infrequent, but large, shifts. We apply the proposed methodology to industry portfolios within a five-factor model setting and show that the threshold Capital Asset Pricing Model (CAPM) provides robust beta estimates coupled with smaller pricing errors compared to the alternative approaches. The results have significant implications for the implementation of smart beta strategies that rely heavily on the accuracy and stability of factor betas and yields.

Firm Characteristics and Global Stock Returns: A Conditional Asset Pricing Model

This paper studies the relation between 36 firm-level characteristics and stock returns in 48 countries using instrumented principal components analysis (IPCA). A non-U.S. country-neutral conditional factor model performs well in describing risk and returns and generates small and statistically insignificant anomaly intercepts when allowing for three or more latent factors. The non-U.S. model performs better in emerging than in developed markets, while showing substantial differences across countries. On average, only ten characteristics contribute significantly to the models' performance. Market beta, momentum and firm size characteristics instrument for systemic exposure in U.S. and non-U.S. models, while investment and book-to-market do not.

Modeling Corporate Bond Returns

We propose a conditional factor model for corporate bond returns with five factors and time-varying factor loadings. We have three main empirical findings. First, our factor model excels in describing the risks and returns of corporate bonds, improving over previously proposed models in the literature by a large margin. Second, our benchmark model recommends a systematic bond investment portfolio that significantly outperforms leading corporate credit investment strategies. Third, we find closer integration between debt and equity markets than found in prior literature.

How financially integrated are trading blocs in Africa?

This paper suggests a new methodology to measure the extent of financial integration between countries or regions based on the combination of the dynamic conditional correlation-general autoregressive conditional heteroscedasticity (DCC-GARCH) and factor models. Unlike previous methodologies, this proposed methodology is able to assess whether countries decouple from the regional factor. The methodology is applied to assess the degree of financial integration between three African trading blocs, namely, the Common Market of Eastern and Southern Africa (COMESA), the East African Community (EAC) and the Southern African Development Community (SADC). The results of the empirical analysis show that not all countries within each of the three trading blocs are integrated to their regional factors. Moreover, the finding of the paper shows a ‘decoupling’ between some of the trading blocs.

Currency carry trades and the conditional factor model

This study employs a conditional factor model in order to investigate the time-varying profitability of currency carry trades. To that end, I estimate conditional alphas and betas on the popular dollar and carry factors through the use of a nonparametric approach. The empirical results illustrate that the alphas and betas vary over time. Furthermore, I find that the alpha of a high interest rate currency portfolio increases in a trough in a business cycle and in a state of high market uncertainty. However, the beta on the dollar factor decreases in these market conditions, suggesting that investors reduce the foreign currency risk exposure.

Currency Carry Trades and the Conditional Factor Model

This study employs a conditional factor model in order to investigate the time-varying profitability of currency carry trades. To that end, I estimate conditional alphas and betas on the popular dollar and carry factors through the use of a nonparametric approach. The empirical results illustrate that the alphas and betas vary over time. Furthermore, I find that the alpha of a high interest rate currency portfolio increases in a trough in a business cycle and in a state of high market uncertainty. However, the beta on the dollar factor decreases in these market conditions, suggesting that investors reduce the foreign currency risk exposure.

Estimation of Large Dimensional Conditional Factor Models in Finance

This chapter provides an econometric methodology for inference in large-dimensional conditional factor models in finance. Changes in the business cycle and asset characteristics induce time variation in factor loadings and risk premia to be accounted for. The growing trend in the use of disaggregated data for individual securities motivates our focus on methodologies for a large number of assets. The beginning of the chapter outlines the concept of approximate factor structure in the presence of conditional information, and develops an arbitrage pricing theory for large-dimensional factor models in this framework. Then we distinguish between two different cases for inference depending on whether factors are observable or not. We focus on diagnosing model specification, estimating conditional risk premia, and testing asset pricing restrictions under increasing cross-sectional and time series dimensions. At the end of the chapter, we review some of the empirical findings and contrast analysis based on individual stocks and standard sets of portfolios. We also discuss the impact on computing time-varying cost of equity for a firm, and summarize differences between results for developed and emerging markets in an international setting.

New Evidence on Conditional Factor Models

We estimate conditional multifactor models over a large cross section of stock returns matching 25 CAPM anomalies. Using conditioning information associated with different instruments improves the performance of the Hou, Xue, and Zhang (HXZ) (2015) and Fama and French (FF) (2015), (2016) models. The largest increase in performance holds for momentum, investment, and intangibles-based anomalies. Yet, there are significant differences in the performance of scaled models: HXZ clearly dominates FF in explaining momentum and profitability anomalies, while the converse holds for value–growth anomalies. Thus, the asset pricing implications of alternative investment and profitability factors (in a conditional setting) differ in a nontrivial way.

Change‐Point Detection in the Conditional Correlation Structure of Multivariate Volatility Models

ABSTRACTWe propose semiparametric CUSUM tests to detect a change-point in the correlation structures of nonlinear multivariate models with dynamically evolving volatilities. The asymptotic distributions of the proposed statistics are derived under mild conditions. We discuss the applicability of our method to the most often used models, including constant conditional correlation (CCC), dynamic conditional correlation (DCC), BEKK, corrected DCC, and factor models. Our simulations show that, our tests have good size and power properties. Also, even though the near-unit root property distorts the size and power of tests, de-volatizing the data by means of appropriate multivariate volatility models can correct such distortions. We apply the semiparametric CUSUM tests in the attempt to date the occurrence of financial contagion from the US to emerging markets worldwide during the great recession. Supplementary materials for this article are available online.