Abstract
In this chapter we fit univariate and bivariate time series models in the tradition of Box and Jenkins (1976) and Granger and Newbold (1977) and apply traditional Granger causality testing following the Ashley et al. (1980) methodology. Second, we estimate Vector Autoregressive Models (VAR) and Chen and Lee (1990) Vector ARMA (VARMA) causality test. We test two series for causality: (1) stock prices and mergers and (2) the money supply and stock prices.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
One must apparently be even more careful with the Box–Pierce test on sums of squared ρ k .
- 2.
OLS estimation suffices to produce unbiased estimates, since all the bivariate models considered are reduced forms. It also allows one to consider variants of one equation without disturbing the forecasting results from the other, and it is computationally simpler. On the other hand, where substantial contemporaneous correlation occurs between the residuals, seemingly unrelated regression GLS estimation can be expected to yield noticeably better parameter estimates and post-sample forecasts. All estimation in this study is OLS; a re-estimation of our final bivariate model using GLS might strengthen our conclusions somewhat.
- 3.
Alternatively, one might fit both models to the sample period, produce forecasts of the first post-sample observation, reestimate both models with that observation added to the sample, forecast the second post-sample observation, and so on until the end of the post-sample period. This would, of course, be more expensive than the approach in the text.
- 4.
If one finds that one model (using a wider information set, say) fits better than another, one is really saying “If I had known that at the beginning of the sample period, I could have used that information to construct better forecasts during the sample period.” But this is not strictly operational and thus seems somewhat contrary in spirit to the basic definition of causality that we employ.
- 5.
The merger history of the United States was studied by Nelson (1959), who reported that mergers were highly correlated with stock prices and industrial production from 1895 to 1954. Nelson (1966) later found that stock prices lead mergers by over 5 months (5.25) over the 1919–1961 period. Melicher et al. (1983) and Guerard (1985) used ARIMA and transfer function modeling to find that stock prices lead mergers. Guerard and McDonald (1995) reported that the annual merger series from 1895 to 1979 was a near-random walk and that outlier-estimated time series models did not statistically outperform the naïve random walk with drift model. Golbe and White (1993) fit a sine wave to a “spliced” US annual merger history and found that a sine wave, representing a 40-year merger model, described the behavior of mergers.
- 6.
The US merger history was characterized by George Stigler (1950) to have occurred in three waves. The first major merger movement began in 1879, with the creation of the Standard Oil Trust, and ended with the depression of 1904. During the merger movement, giant corporations were formed by the combination of numerous smaller firms. The smaller companies represented nearly all the manufacturing or refining capacity of their industries. The forty largest firms in the oil-refining industry, comprising over ninety percent of the country’s refining capacity and oil pipelines for its transportation, combined to form Standard Oil. In the two decades following the rise of Standard Oil, similar horizontal mergers created single dominant firms in several industries. These dominant firms included the Cottonseed Oil Trust (1884), the Linseed Oil Trust (1885), the National Lead Trust (1887), the Distillers and Cattle Feeders (1887), and the Sugar Refineries Company (1887). The trust form of organization was outlawed by court decisions. But merger activities continued to create “near” monopolies as the single corporation or holding company organization became dominant. The Diamond Match Company (1889), the American Tobacco Company (1890), the United States Rubber Company (1892), the General Electric Company (1892), and the United States Leather Company (1893) were created by the development of the modern corporation or holding company.
The height of the merger movement was reached in 1901 when 785 plants combined to form America’s first billion-dollar firm, the United States Steel Corporation. The series of mergers creating the US Steel allowed it to control 65 % of the domestic blast furnace and finished steel output. This growth in concentration was typical of the first merger movement. The early mergers saw 78 of 92 large consolidations gain control of 50 % of their total industry output, and 26 secure 80 % or more.
The first major merger movement occurred during a period of rapid economic growth. The economic rationale for the large merger movement was the development of the modern corporation, with its limited liability, and the modern capital markets, which facilitated the consolidations through the absorption of the large security issues necessary to purchase firms. Nelson found that the mergers were highly correlated to the period’s stock prices and industry production. However, mergers were more sensitive to stock prices. The expansion of security issues allowed financiers the financial power necessary to induce independent firms to enter large consolidations. The rationale for the first merger movement was not one of trying to preserve profits despite slackening demand and greater competitive pressures. Nor was the merger movement the result of the development of the national railroad system, which reduced geographic isolation and transportation costs. The first merger movement ended in 1904 with a depression, the onset of which coincided the Northern Securities case. Here it was held, for the first time, that antitrust laws could be used to attack mergers leading to market dominance.
A second major merger movement stirred the country from 1916 to the depression of 1929. This merger movement was only briefly interrupted by the First World War and the recession of 1921 and 1922. The approximately 12,000 mergers of the period coincided with the stock market boom of the 1920s. Although mergers greatly affected the electric and gas utility industry, market structure was not as severely concentrated by the second movement as it was by the first merger movement. Stigler (1950) concluded that mergers during this period created oligopolies, such as Bethlehem Steel and Continental Can. Mergers, primarily vertical and conglomerate in nature as opposed to the essentially horizontal mergers of the first movement, did affect competition adversely. The conglomerate product-line extensions of the 1920s were enhanced by the high-cross elasticities of demand for the merging companies’ products. Antitrust laws, though not seriously enforced, prevented mergers from creating a single dominant firm. Merger activity diminished with the depression of 1929 and continued to decline until the 1940s.
The third merger movement began in 1940; mergers reached a significant proportion of firms in 1946 and 1947. The merger action from 1940 to 1947, although involving 7.5 % of all manufacturing and mining corporations and controlling 5 % of the total assets of the firms in those industries, was quite small compared to the merger activities of the 1920s. The mergers of the 1940s included only one merger between companies with assets exceeding 50 million dollars and none between firms with assets surpassing 100 million dollars. The corresponding figures for the mergers of the 1920s were 14 and eight, respectively. Eleven firms acquired larger firms during the mergers of the 1920s than the largest firm acquired during the 1940s merger. The mergers of the 1940s affected competition far less than did the two previous merger movements, with the exception of the food and textile industries. The acquisitions by the large firms during the 1940s rarely amounted to more than seven percent of the acquiring firms’ 1939 assets or to as much as a quarter of ~ the acquiring firm's growth rate from 1940 to 1947. Approximately 5 billion dollars of assets were held by acquired or merged firms over the 1940–1947 period. Smaller firms were generally acquired by larger firms. Companies with assets exceeding 100 million dollars acquired, on average, firms with assets of less than two million dollars. The larger firms tended to engage in a greater number of acquisitions than smaller firms. The acquisitions by the larger, acquiring firms tended to involve more firms than did those acquired by smaller, acquiring firms. Mergers added relatively less to the existing size of the larger acquiring firms in the early period of the third merger movement. The relatively smaller asset growth of the larger acquiring firms is in accordance with the third merger movement’s generally small effects on competition and concentration. One factor contributing to the maintenance of competition was the initiative for the mergers coming from the owners of the smaller firms. Financiers and investment bankers did not play a prominent part in the early third merger movement, but certainly have in the 1992–2011 period.
- 7.
Neither stock prices nor LEI passed the AGS (1980) causality test for mergers.
- 8.
The SCA outlier estimation using stock prices as the input series is:
- 9.
Had one modeled stock prices and mergers for the 1979–2011 period, one finds only a contemporaneous relationship and no strong causality findings.
- 10.
We use M1P, a variation on M1, rather than M3, that was used in the earlier studies because M3 was discontinued in the FRED database.
References
Ashley, R.A. 2003. “Statistically Significant Forecasting Improvements: How Much Out-of-sample Data is Likely Necessary?” International Journal of Forecasting 19, 229-240.
Ashley, R. 1998. “A New Technique for Postsample Model Selection and Validation.” Journal of Economic Dynamics and Control 22, 647-665.
Ashley, R. A., C. W. J., Granger, and R. L. Schmalensee. 1980. “Advertising and Aggregate Consumption: An Analysis of Causality”. Econometrica 48, 1149-1168.
Box, G.E.P. and Jenkins, G.M. 1976. Time Series Analysis Forecasting and Control. San Francisco: Holden-Day.
Butters, J. K., J. Lintner, and W. L. Cary. 1951. Corporate Mergers. Boston: Division of Research, Graduate School of Business, Harvard University Press.
Chen, C. and Lee, C.J. (1990). “A VARMA Test on the Gibson Paradox.” Review of Economics and Statistics 72: 96-107.
Clements, M. P. and D. F. Hendry. 1998. Forecasting Economic Time Series. Cambridge: Cambridge University Press.
Doornik, J. A. and D. F. Hendry. Empirical Econometric Modelling. Timberlake Consultants, Ltd.
Doornik, J. A. and D. F. Hendry. Modelling Dynamic Systems. Timberlake Consultants, Ltd.
Feige, E.L. and D. K. Pearce, 1976. “Economically Rational Expectations: Are Innovations in the Rate of Inflation Independent of Innovations in Measures of Monetary and Fiscal Policy?” Journal of Political Economy 84: 499-522.
Granger, C.W.J. 1969. “Investigating Causal Relations by Econometric Models and Cross-Spectral Methods.” Econometrica 37: 424-438.
Granger, C.W.J. and P. Newbold. 1974. “Spurious Regressions in Econometrics.” Journal of Econometrics 1: 111-120.
Granger, C.W.J. and P. Newbold. 1977. Forecasting Economic Time Series. New York: Academic Press.
Guerard, J.B., Jr. 1985. “Mergers, Stock Prices, and Industrial Production: An Empirical Test of the Nelson Hypothesis”, in O.D. Anderson, Ed., Time Series Analysis: Theory and Practice 7, Amsterdam: Elsevier, 239-248.
Golbe, D. L. and L. J. White. 1993. ”Catch a Wave: The Time Series Behavior of Mergers”. The Review of Economics and Statistics 75, 493-499.
Guerard, J. B., Jr. and J. B. McDonald. 1995. “Mergers in the United States: A Case Study in Robust Time Series”, in C. F. Lee., Ed., Advances in Financial Planning and Forecasting, Greenwich, CT: JAI Press, 95–107.
Guerard, John B., Jr. 2004. “The Forecasting Effectiveness of the U.S. Leading Economic Indicators: Further Evidence and Initial G7 Results.” In P. Dua, Ed., Business Cycles and Economic Growth: An AnalysisUsing Leading Indicators. New York: Oxford University Press, 174-187.
Hamilton, J.D. 1994. Time Series Analysis, Princeton, Princeton University Press.
Hoover, K.D. 2001. Causality in Macroeconomics. Cambridge: Cambridge University Press.
Lee, C.J. and E. Petruzzi, E. 1986. “The Gibson Paradox and Monetary Standard.” The Review of Economics and Statistics 68: 189-196.
Levanon G., A. Ozyildirim, B. Schaitkin, and J. Zabinska. 2011. “Comprehensive Benchmark Revisions for the conference Board Leading Economic Index for the United States.” The Conference Board. EPWP # 1–06. December 2011.
Liu, L.M. 1999. Forecasting and Time Series Analysis Using the SCA Statistical System: Volume 2. Chicago: Scientific Computing Associates Corp.
Lutkepohl, H. 1993. Introduction to Multiple Time Series Analysis. 2nd edition. Berlin: Springer-Verlag.
McCracken, M. 2000. “Robust Out-of-Sample Inference.” Journal of Econometrics 39, 195-223
Melicher, R., W. J. Ledolter, and L. J. D’Antonio. 1983. “A Time Series Analysis of Aggregate Merger Activity”. The Review of Economics and Statistics 65, 423-430.
Nelson, C.R. and G.B. Schwertz, G.B. 1982. “Tests for Predictive Relationships Between Time Series Variables: A Monte Carlo Investigation.” Journal of the American Statistical Association 77: 11-18.
Nelson, R. L. 1959. Merger Movements in American Industry, 1895–1956. Princeton: Princeton University Press.
Nelson, R. L. 1966. “Business Cycle Factors in the Choice between Internal and External Growth”, in W.W. Alberts and J. E. Segall, Eds, The Corporate Merger. Chicago: The University of Chicago Press, 52-66.
Pierce, D.A. and L. D. Haugh, L.D. 1977. “Causality in Temporal Systems: Characterizations and Survey,” Journal of Econometrics 5: 265-293.
Sargent, T.J. 1973. “Interest Rates and Prices in the Long Run.” Journal of Money, Credit and Banking: 385-463.
Shiller, R.J. and Siegel, J.J. 1997. “The Gibson Paradox and Historical Movement in Real Interest Rates.” Journal of Political Economy 85: 891-907.
Sims, C.A. 1972. “Money, Income and Causality,” American Economic Review 62: 540-552.
Sims, C.A. 1980. “Macro-Economics and Reality.” Econometrica 48: 1-48.
Stigler, G. J. 1950. Monopoly and Oligopoly by Merger. American Economic Review 40, 23-34.
Tiao, G.C. and G.E. P. Box, G.E.P. 1981. “Modeling Multiple Time Series with Applications.” Journal of American Statistical Association 76: 802-816.
Tiao, GC. and R.S. Tsay. 1983, “Multiple Time Series Modeling and Extended Sample Cross Correlations,” Journal of Business and Economic Statistics 1: 43–56.
Thomakos, D. and J. Guerard. 2004. “Naïve, ARIMA, Transfer Function, and VAR Models: A Comparison of Forecasting Performance.” The International Journal of Forecasting 20, 53-67.
Wicksell, K. 1907. “The Influence of the Rate of Interest on Prices.” Economic Journal 17: 213-220.
Zellner, A. and Palm, F. 1974. “Time Series Analysis and Simultaneous Equation Econometric Models.” Journal of Econometrics 2: 17-54.
Zarnowitz, V. 1992. Business Cycles: Theory, History, Indicators, and Forecasting. Chicago: The University of Chicago Press.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer Science+Business Media New York
About this chapter
Cite this chapter
Guerard, J.B. (2013). Transfer Function Modeling and Granger Causality Testing. In: Introduction to Financial Forecasting in Investment Analysis. Springer, New York, NY. https://doi.org/10.1007/978-1-4614-5239-3_5
Download citation
DOI: https://doi.org/10.1007/978-1-4614-5239-3_5
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4614-5238-6
Online ISBN: 978-1-4614-5239-3
eBook Packages: Business and EconomicsEconomics and Finance (R0)