skip to content
 

 


 

  • "Arbitrage Pricing Theory, the Stochastic Discount Factor and Estimation of Risk Premia from Portfolios", by M. Hashem Pesaran and Ron P. Smith, forthcoming in Econometrics and Statistics, November 2021, CESifo Working Paper No. 9001

    Abstract: The arbitrage pricing theory (APT) attributes differences in expected returns to exposure to systematic risk factors. Two aspects of the APT are considered. Firstly, the factors in the statistical asset pricing model are related to a theoretically consistent set of factors defined by their conditional covariation with the stochastic discount factor (SDF) used to price securities within inter-temporal asset pricing models. It is shown that risk premia arise from non-zero correlation of observed factors with SDF and the pricing errors arise from the correlation of the errors in the statistical model with SDF. Secondly, the estimates of factor risk premia using portfolios are compared to those obtained using individual securities. It is shown that in the presence of pricing errors consistent estimation of risk premia requires a large number of not fully diversified portfolios. Also, in general, it is not possible to rank estimators using individual securities and portfolios in terms of their small sample bias.
    JEL Classifications: C38, G12
    Key Words: Arbitrage Pricing Theory, Stochastic Discount Factor, portfolios, factor strength, identification of risk premia, two-pass regressions, Fama-MacBeth.
    Full Text: https://www.econ.cam.ac.uk/people-files/emeritus/mhp1/fp21/PSonPortfolios_15_November_2021.pdf
    CESifo Full Text: https://www.cesifo.org/node/62812

     

  • "An Augmented Anderson-Hsiao Estimator for Dynamic Short-T Panels", by Alexander Chudik and M. Hashem Pesaran, forthcoming in Econometric Reviews, July 2021, CESifo WP no. 6688.

    Abstract: This paper introduces the idea of self-instrumenting endogenous regressors in settings when the correlation between these regressors and the errors can be derived and used to bias-correct the moment conditions. The resulting bias-corrected moment conditions are less likely to be subject to the weak instrument problem and can be used on their own or in conjunction with other available moment conditions to obtain more efficient estimators. This approach can be applied to estimation of a variety of models such as spatial and dynamic panel data models. This paper focuses on the latter, and proposes a new estimator for short T dynamic panels by augmenting Anderson and Hsiao (AAH) estimator with bias-corrected quadratic moment conditions in first differences which substantially improve the small sample performance of the AH estimator without sacrificing on the generality of its underlying assumptions regarding the fixed effects, initial values, and heteroskedasticity of error terms. Using Monte Carlo experiments it is shown that AAH estimator represents a substantial improvement over the AH estimator and more importantly it performs well even when compared to Arellano and Bond and Blundell and Bond (BB) estimators that are based on more restrictive assumptions, and continues to have satisfactory performance in cases where the standard GMM estimators are inconsistent. Finally, to decide between AAH and BB estimators we also propose a Hausman type test which is shown to work well when T is small and n sufficiently large.
    JEL Classifications: C12, C13, C23
    Key Words: Short-T Dynamic Panels, GMM, Bias-Corrected Moment Conditions, BMM, Self-Instrumenting, Nonlinear Moment Conditions, Panel VARs, Hausman Test, Monte Carlo Evidence.
    Full Text: http://www.econ.cam.ac.uk/emeritus/mhp1/fp21/CP_paper_2021_July16.pdf
    Supplement: http://www.econ.cam.ac.uk/emeritus/mhp1/fp21/CP_online_supplement_2021_July16.pdf
    Data and Codes: http://www.econ.cam.ac.uk/people-files/emeritus/mhp1/fp21/CP_AAH_paper_July_2021_codes_and_data.zip

     

  • "Regional Heterogeneity and U.S. Presidential Elections: Real-Time 2020 Forecasts and Evaluation", by Rashad Ahmed and M. Hashem Pesaran, forthcoming in International Journal of Forecasting, June 2021.

    Abstract: This paper exploits cross-sectional variation at the level of U.S. counties to generate real-time forecasts for the 2020 U.S. presidential election. The forecasting models are trained on data covering the period 2000-2016, using high-dimensional variable selection techniques. Our county-based approach contrasts the literature that focuses on national and state level data but uses longer time periods to train their models. The paper reports forecasts of popular and electoral college vote outcomes and provides a detailed ex post evaluation of the forecasts released in real time prior to the election. It is shown that all of these forecasts outperform autoregressive benchmarks, with a pooled national model using One-Covariate-at-a-time-Multiple-Testing (OCMT) variable selection significantly outperforming all models in forecasting both the U.S. mainland national vote share and electoral college outcomes (forecasting 236 electoral votes for the Republican party compared to 232 realized). This paper also shows that key determinants of voting outcomes at the county level include incumbency effects, unemployment, poverty, educational attainment, house price changes, and international competitiveness. The results are also supportive of myopic voting: economic fluctuations realized a few months before the election tend to be more powerful predictors of voting outcomes than their long-horizon analogues.
    JEL Classifications: C53, C55, D72
    Key Words: Real-time Forecasts, Popular and Electoral College Votes, Simultaneity, High Dimensional Forecasting Models, Lasso, One Covariate at a time Multiple Testing, OCMT.
    Full Text: http://www.econ.cam.ac.uk/people-files/emeritus/mhp1/fp21/AhmedPesaran_Elections_Jun_23_2021.pdf
    Data and Codes: http://www.econ.cam.ac.uk/people-files/emeritus/mhp1/fp21/AhmedPesaran_Replication.zip