Clive Granger was born in Swansea, Wales, in 1934. In his Nobel autobiography Granger takes the self-deprecating view that his life has been conditioned by a series of ‘lucky breaks’, the first of which took his family to Nottingham when he was still a boy and pitched him into school with able and enthusiastic mathematics teachers. He subsequently studied mathematics at Nottingham University, where he was awarded a BA in 1955 and a PhD in 1959. Granger entered university as a student of both economics and mathematics but dropped economics after a year. His PhD was in statistics and his thesis explored economic time series, a topic chosen because it seemed promisingly underdeveloped: Granger could find only one relevant book in the university library (Nobel Foundation, 2004).

Granger’s academic career began in 1955 as a result of another serendipitous event. He had applied for a junior lectureship in statistics at Nottingham, but as a makeweight to save the university the embarrassment of interviewing only one candidate. Not expecting to get the job, he had a relaxed and enjoyable interview; his rival did not and Granger was offered the post (Nobel Foundation, 2004). He was promoted to a lectureship in economics in 1958. In 1959 he was visiting professor at Princeton University, working for a year with Oscar Morgenstern on time series, and in 1963 he was visiting profes sor at Stanford University. Returning to Nottingham, Granger became Reader in Mathematics in 1964 and, from 1965 to 1974, Professor of Applied Statistics and Econometrics. In 1974 he moved as Professor of Economics to the University of California at San Diego (UCSD), a post he still holds. Granger has also been visiting professor at the Australian National University (1977) and visiting fellow of All Souls College, Oxford (1994) and Trinity College, Cambridge (1996).

Granger’s honours and awards include fellowships of the Econometric Society in 1972 and the American Academy of Arts and Sciences in 1994. He became a distinguished fellow of the American Economic Association in 2002, and was president of the Western Economic Association in 2002–03. He became a foreign member of the Finnish Society of Arts and Science in 1997 and a corresponding fellow of the British Academy in 2002. Jointly with Robert Engle, Granger was awarded the 2003 Nobel Memorial Prize in Economics ‘for methods of analysing economic time series with common trends (cointegration)’ (Nobel Foundation, 2004).

Granger’s work on time series has had a paradigmatic influence on macroeconomic research and policy analysis (Hendry, 2004). His development of the concept of cointegration has provided for a hugely fruitful blending of long-run equilibrium macro relationships with short-run perspectives (Nobel Foundation, 2004). The resultant framework is equally satisfactory in terms of economic theory and econometric method.

A key milestone in this work was research done by Granger with his colleague, Paul Newbold. Their 1974 Journal of Econometrics paper, ‘Spurious Regressions in Econometrics’, showed that the application of standard statistical methods to the non-stationary time series that are common in macroeconomics could produce fundamentally misleading results, suggesting the presence of significant relationships between variables when in fact none existed. Nonstationary series exhibit no tendency to return to a given trend or value: short-term disturbances affect longer-term outcomes, as, for example, in the cases of output or employment. Granger (2004, p. 423) notes that his and Newbold’s findings ‘led to a great deal of re-evaluation of empirical work, particularly in macroeconomics, to see if apparent relationships were correct or not. Many editors had to look again at their list of accepted papers’.

A way around the problem of non-stationarity suggested by Granger and Newbold was to pose econometric models using first differences of variables (their rates of increase) rather than their levels (see Hendry, 2004; Nobel Foundation, 2004). This is because first differences of variables are usually stationary – that is they do tend to observe a given trend – and are thus amenable to analysis using standard statistical methods. Spurious results can therefore be avoided. Unfortunately, however, while such an approach is statistically robust in capturing shorter-run economic relationships, it is less satisfactory in terms of economic theory because this usually specifies relationships between the longer-run levels of variables rather than first differences (Nobel Foundation, 2004). Granger’s great contribution was to leap over this seeming impasse between what was desirable from the viewpoint of economic theory and what was necessary for valid statistical method. The key was his discovery that a linear combination of a pair of non-stationary variables may be stationary. This property – which Granger (1981) labelled ‘cointegration’ – facilitates standard statistical analysis but, importantly, it also makes a desirable connection to longer-run economic theorisation – something missing from the purely first differences solution to the problem of non-stationarity just discussed. Cointegration has been widely deployed in macroeconomic research and policy making using the ‘error correction model’ first proposed by Dennis Sargan (Granger, 2004). With Andrew Weiss, Granger also devised a test for cointegration between non-stationary variables (Granger and Weiss, 1983). Finally, in what Diebold (2004, p.166) has called ‘perhaps the most cited paper in the history of econometrics’, Granger, in collaboration with his fellow Nobel Laureate Robert Engle (Engle and Granger, 1987), provided a two-step estimation procedure for cointegrated variables.

Other notable work by Granger includes a book on spectral analysis (written in association with Michio Hatanaka; see Granger and Hatanaka, 1964); and, with J.M. Bates, research on the superiority of pooled forecasts (Bates and Granger, 1969a). His associations with Oscar Morgenstern and Paul Newbold resulted in two books – Granger and Morgenstern (1970) and Granger and Newbold (1977); and he has also edited a volume with Robert Engle on cointegration (Engle and Granger, 1991). In Granger (1969b) he offered an influential interpretation of causality – now referred to as ‘Granger causality’ – and he has also undertaken important research with Roselyn Joyeux, among others, on long-memory models (see, for example, Granger and Joyeux, 1980). With Timo Terasvirta he has written an overview of non-linear time-series modelling (Granger and Terasvirta, 1993). Finally, among Granger’s most recent work is his contribution to a multi-author study of the future of the Amazon rainforest in Brazil (Granger et al., 2003).

**Main Published Works**

(1964), Spectral Analysis of Economic Time Series (in association with M. Hatanaka), Princeton, NJ: Princeton University Press.

(1969a), ‘The Combination of Forecasts’ (with J.M. Bates), Operations Research Quarterly, 20, pp. 451–68.

(1969b), ‘Investigating Causal Relationships by Econometric Models and Cross-Spectral Methods’, Econometrica, 37, July, pp. 424–38.

(1970), Predictability of Stock Market Prices (with O. Morgenstern), Lexington, MA: D.C. Heath.

(1974), ‘Spurious Regressions in Econometrics’ (with P. Newbold), Journal of Econometrics, 2, July, pp. 111–20.

(1977), Forecasting Economic Time Series (with P. Newbold), New York: Academic Press, 2nd edn 1986.

(1980), ‘An Introduction to Long-Memory Time Series Models and Fractional Differencing’ (with R. Joyeux), Journal of Time Series Analysis, 1, pp. 15–30.

(1981), ‘Some Properties of Time Series Data and Their Use in Econometric Model Specification’, Journal of Econometrics, 16, May, pp. 121–30.

(1983), ‘Time Series Analysis of Error Correction Models’ (with A. Weiss), in S. Karlin, T. Amemiya and L.A. Goodman (eds), Studies in Econometrics, Time Series, and Multivariate Statistics, New York: Academic Press, pp. 255–78.

(1987), ‘Co-integration and Error-Correction: Representation, Estimation and Testing’ (with R.F. Engle), Econometrica, 55, March, pp. 251–76. (1991), Long-Run Economic Relationships. Readings in Cointegration (ed. with R.F. Engle), Oxford: Oxford University Press.

(1993), Modelling Nonlinear Economic Relationships (with T. Terasvirta), Oxford: Oxford University Press.

(2003), The Dynamics of Deforestation and Economic Growth in the Brazilian Amazon (with L. Andersen, E. Reis, D. Weinhold and S. Wunder), Cambridge: Cambridge University Press.

(2004), ‘Time Series Analysis, Cointegration and Applications’, American Economic Review, 94, June, pp. 421–5.

**Secondary Literature**

Diebold, F.X. (2004), ‘The Nobel Memorial Prize for Robert F. Engle’, Scandinavian Journal of Economics, 106 (2), pp. 165–85.

Hendry, D.F. (2004), ‘The Nobel Memorial Prize for Clive W.J. Granger’, Scandinavian Journal of Economics, 106 (2), pp. 187–213.

Royal Swedish Academy of Sciences (2004), ‘The Nobel Memorial Prize in Economics 2003’, Scandinavian Journal of Economics, 106 (2), pp. 163–4.