Abstract we study biases associated with regression models in which persistent lagged variables predict stock returns, either linearly or in interaction with contemporaneous values of a market. Our simulations suggest that many of the regressions in the literature, based on individual predictor variables, may be spurious suggested citation. Econometricians have known about type ii spurious regression since yule. Total downloads of all papers by sergei sarkissian. Citeseerx document details isaac councill, lee giles, pradeep teregowda. A simple solution of the spurious regression problem ideasrepec. This is true even when the mea sured regressor is highly persistent.
Estimating a linear regression on two variables can be visualised as fitting a line. He taught in britain, at the university of nottingham and in the united states, at the university of california, san diego. This is one of the most common mistakes people make in the world of statistics. Lets see what is the problem, and how can we fix it. He has published research in the journal of economics and business, the journal of financial markets, the journal of finance, the financial analysts journal, the journal of investment management, has several book chapters, and has won the best investments paper award. This paper introduces a new approach to the modelling of a stationary long run component, which is an autoregressive process with near unit root and small sigma innovation. Abstract even though stock returns are not highly autocorrelated, there is a spurious regression bias in predictive regressions for stock. Islamic banking finance, monetary economics and econometrics. When looking at multiple time series, one can often be fooled by spurious correlations. Econometrics is the application of statistical methods to economic data in order to give. We confirm this with additional simulations, not reported in the tables, where we set p 0 and.
Spurious regressions in financial economics 2 results will differ from those in the classical setting. Newbold, regressions in econometrics to take is that of the. However, formatting rules can vary widely between applications and fields of interest or study. Simin abstract even though stock returns are not highly autocorrelated, there is a spurious regression bias in predictive regressions for stock returns related to the classic studies of yule 1926 and granger and newbold 1974. The effects of data snooping and spurious regression, journal of financial and quantitative analysis, 2007. Some important applications of spurious regressions in economics and finance, although this list is by no means exhaustive, include plosser. Ferson, sarkissian, and simin 2003a, b, 2009 noted that their finding of spurious regression in financial economics is well outside the classical setting of yule 1926 and granger and newbold 1974, in part because the dependent variable in the predictive regression, i. Go to the next page of charts, and keep clicking next to get through all 30,000. Understanding spurious regression in financial economics understanding spurious regression in financial economics deng, ai 20140119 00.
A wellknown case of a spurious relationship can be found in the timeseries literature, where a spurious regression is a regression that provides misleading statistical evidence of a linear relationship between independent nonstationary variables. Introduction it is very common to see reported in applied econometric literature time series. Why do we sometimes get nonsense correlations between timeseries. Newboldsome comments on the evaluation of economic forecasts. As was pointed out by granger and newbold 1974, the levels of many economic timeseries are integrated or nearly so, and if these data are used in a regression model then a high value for the coefficient of determination r 2 is likely to arise, even when the. We find that the regressions are reasonably well specified for conditional betas, even in settings where simple predictive regressions are severely biased. In this case, the usual statistical results for the linear regression model hold. Giles department of economics university of victoria, b. Spurious regression and data mining in conditional asset.
The specification, estimation, diagnostic testing, and practical usage of dynamic models for economic and financial time series present a host of unique challenges, requiring the. The odds for a spurious regressor become even longer when considering the second hypothesis, as we see from the second column of table 1. Asset pricing program even though stock returns are not highly autocorrelated, there is a spurious regression bias in predictive regressions for stock returns related to the classic studies of yule 1926 and granger and newbold. This work should prove useful to researchers in statistics, timeseries econometrics and applied economics. This paper studies the estimation of asset pricing model regressions with conditional alphas and betas, focusing on the joint effects of data snooping and spurious regression. If the standard instruments employed in the literature arise as the result of a collective search through. However, even a small rsquared can signal economically significant predictability. Moreover in the presence of a long run component, spurious regressions arise. Spurious regressions in financial economics wayne e ferson. Financial econometrics, mathematics and statistics.
National tsing hua university, department of quantitative finance, hsinchu city. Data mining for predictor variables interacts with spurious regression bias. Spurious regressions in financial economics 5 fraction of the variance in stock returns. Even though stock returns are not highly autocorrelated, there is a spurious regression bias in predictive regressions for stock returns related to the classic studies of yule 1926 and granger and newbold 1974. Indeed, spurious regressions have played a fundamental role in the building of modern time series econometrics and have revolutionized many of the procedures used in applied. Spinoffs from this research range from unitroot tests to cointegration and errorcorrection models. In 2003, granger was awarded the nobel memorial prize in economic sciences, in recognition of. Book tomarket, dividend yield, and expected market returns. We show that a combination of a noise and a long run component can explain the long run predictability puzzle pointed out in famafrench 1988.
Spurious regression with stationary time series msr economics. Understanding spurious regression in financial economics. Written by leading academics in the quantitative finance field, this book allows readers to implement the principles behind financial econometrics and statistics through realworld applications and problem sets. Wayne ferson, john merrick 1987 nonstationarity and the stage of the business cycle effects in consumptionbased asset pricing relations, journal of financial economics 18, 127146. Newbold university of nottingham, nottingham ng7 zrd, england received may 1973, revised version received december 1973 1. This book provides an overview of results about spurious regression, pulled from disperse sources, and explains their implications.
Or for something totally different, here is a pet project. Even though stock returns are not highly autocorrelated, there is a spurious regression bias in predictive regressions for stock returns related to the classic. Spurious regressions in econometrics essays in econometrics. Even though stock returns are not highly autocorrelated, there is a spurious regression bias in predictive regressions for stock returns related to the classic studies of yule 1926 and granger. Inference for noisy long run component process munich. When p 0, and there is no persistence in the true expected return, the spur ious regression phenomenon is not a concern. Spurious regressions springerlink skip to main content.
For the first threequarters of the 20th century the main workhorse of applied econometrics was the basic regression. Newbold university of nottingham, nottingham ng7 2rd, england received may 1973, revised version received december 1973 1. The spurious regression phenomenon in least squares occurs for a wide range of data generating processes, such as driftless unit roots, unit roots with drift, long memory, trend and brokentrend stationarity. We focus on the issue of spurious regression, related to the classic studies of yule 1926 and granger and newbold 1974. Asset pricing models with conditional betas and alphas. Even though stock returns are not highly autocorrelated, there is a spurious regression bias in predictive regressions for stock returns related to the classic studies of yule 1926 and. Some important applications of spurious regressions in economics and finance, although this list is by no means exhaustive, include plosser et al. When is the next time something cool will happen in space. The second issue is data mining, as studied for stock returns by lo and mackinlay 1990, foster, smith and whaley 1997 and others. That hypothesis is supported as strongly by x t as it is by s t only once in every 105,000 series.
An introductory economics textbook describes econometrics as allowing. A simple solution of the spurious regression problem in. The title indicates the terms spurious regression was known at least as early as. Wayne ferson, shmuel kandel, robert stambaugh 1987 tests of asset pricing with timevarying expected risk premiums and market betas, journal of finance. It will appeal to a lessserved market of advanced students and scholars in finance, economics, accounting, and statistics. Since type i spurious regression has not been pointed out in the literature, spurious regression of type ii is the only one usually discussed in econometrics.
Canada abstract a spurious regression is one in which the timeseries variables are nonstationary and independent. The effects of data snooping and spurious regression, journal of financial and quantitative analysis, cambridge university. Economic fluctuations and growth, monetary economics, technical working papers socalled spurious regression relationships between randomwalk or strongly autoregressive variables are generally accompanied by clear signs of severe autocorrelation. Spurious regressions and cointegration karl whelan school of economics, ucd february 22, 2011 karl whelan ucd spurious regressions and cointegration february 22, 2011 1. For example, kandel and stambaugh 1996 and fleming, kirby and ostdiek 2001 find that optimal portfolios respond by a substantial amount to small rsquares in standard models. I am using morgan stanley ms symbol for illustration, precrisis time continue reading. Spurious regression problem dates back to yule 1926. The two effects reinforce each other, because more highly persistent series are more. Econometrics for financial and macroeconomic time series. If some or all of the variables in the regression are. We find that spurious regression is a concern in regressions of stock returns on persistent lagged instruments, especially when the predictable component of returns is large. Deng 2005 recommends lag length fixed constants b in the interval.
163 1146 995 158 522 561 1395 343 70 110 977 64 819 978 880 1074 191 1407 1352 64 122 288 1279 310 1110 509 564 72 1173 491 582 1519 1519 40 1049 842 232 984 960 727 388 710 610 451