Perspectives on Econometrics and Applied Economics

A Tribute to Sir Clive Granger

Mark Taylor

• Éditeur: Routledge
• Année: 2014
• ISBN:
  9781317978497
• Édition: 1
• Sujet: Business
• Sous-sujet: Business General

This volume is dedicated to the memory and the achievements of Professor Sir Clive Granger, economics Nobel laureate and one of the great econometricians and applied economists of the twentieth and early twenty-first centuries. It comprises contributions from leading econometricians and applied economists who knew Sir Clive and interacted with him over the years, and who wished to pay tribute to him as both a great economist and econometrician, and as a great man.

This book was originally published as a special issue of Applied Financial Economics.

Special issue in honour of Clive Granger

Mark P. Taylor

Editor, Applied Economics Series

This special issue of Applied Financial Economics is dedicated to the memory and the achievements of Professor Sir Clive Granger, economics Nobel laureate and one of the great econometricians and applied economists of the twentieth and early twenty-first centuries. As editor of the Applied Economics journals I am proud that Sir Clive had such a long and distinguished association with the journals; indeed he was one of the early editors of Applied Economics and was for many years on the editorial board of both Applied Economics and Applied Financial Economics. Sir Clive also published a number of his own papers in the journals, the first in 1971 (in Applied Economics) and the last in 2001 (in Applied Financial Economics).

In September 2008, we held a conference at Clare College, Cambridge to celebrate the fortieth anniversary of the founding of the Applied Economics series of journals, and we were honoured that Sir Clive attended the conference and gave the keynote address, which was a personal reflection on the Applied Economics journals, including the early history, the text of which is reproduced as the first article in this special issue.

The remainder of the special issue is composed of contributions from leading econometricians and applied economists who knew Sir Clive and interacted with him over the years and wished to pay tribute to him as both a great economist and econometrician and as a great man.

As well as Sir Clive’s professional distinction, those who had the privilege to interact with him in person will remember him as a kind and unassuming man who, although he reached the highest echelons of our profession, was never too busy to exchange ideas or correspond on a research paper or just pass the time of day at a conference or seminar. In my own case, I well remember meeting Sir Clive for the first time at a seminar he gave at the London School of Economics in 1985, where I first heard Sir Clive explain the concept of cointegration. Having by chance found myself sharing the very slow lift (elevator in US parlance) from the LSE foyer with a middle aged man with silver hair (mine was dark in those days), the man struck up a conversation with me and spent several minutes quizzing me about my research with great animation. It was only when we were entering the seminar room that he suddenly shook my hand and said, ‘Oh, by the way, I’m Clive Granger, really nice to meet you’. It sounds a little prosaic now, but I was dumbfounded to find out who my interlocutor was and quite amazed that the great man would have had so great an interest in a young researcher whom he had never met before. Over the years, however, I observed that Sir Clive was always kind and generous with his time towards younger researchers: Sir Clive could walk with the leaders of our profession, yet never lost the common touch.

In the first of the contributions to the special issue in honour of Sir Clive, Andrew Harvey examines the relationship between inflation and the output gap (i.e. the Phillips curve), and finds that it can be modelled simply and effectively using either single equation or bivariate unobserved components models. The fitted models consistently reveal that an output gap 2% above trend is associated with an annual inflation rate that is 1% above core inflation.

In his contribution in tribute to Sir Clive, David Hendry revisits one of his own seminal contributions to applied economics, namely the equilibrium-correction modelling of aggregate real consumers’ expenditure, using Autometrics. The many selection decisions involved in developing viable empirical models are discussed in a setting where there are more candidate explanatory variables than observations. Additional tests of the selected model include whether it encompasses the original specification, evidence of non-linearity, and the exogeneity of the conditioning variables.

Ken Wallis’s paper revisits two of Sir Clive’s important papers on forecast combination. In particular, Wallis returns to important topics in this literature, namely the impact of forecasters’ different information sets on the original point forecast combination result and properties of different methods for combining density forecasts. In the context of the latter, Wallis establishes that logarithmic combination has some advantage over linear combination: a result which parallels Sir Clive’s important finding of the inefficiency of mean forecasts.

In their contribution in tribute to Sir Clive, Nii Ayi Armah and Norman R. Swanson contribute to the growing literature on macroeconomic forecasting and the use of diffusion indices to capture the salient features of large data sets. In particular, they compare the predictive content for output growth and inflation of selected variables with that of a group of observable factor proxies and diffusion indices based on large data sets. They find that a judiciously chosen set of macroeconomic indicators can effectively provide the same macroeconomic policy-relevant information as that contained in a large-scale time series dataset—although the large-scale datasets are still required in order to select the key indicator variables or confirm the prior choice of key variables. Armah and Swanson also find that the yield spread between Treasury or corporate yields and the federal funds rate are extremely useful in generating macro-economic forecasts.

Philip Hans Franses and Heleen Mees, in their contribution, examine an interesting hypothesis concerning Chinese real GDP growth, namely that news on GDP should not impact stock market fluctuations or their volatility if it is known to traders that the Chinese government manipulates GDP statistics for political reasons.

In their paper, Timo Teräsvirta and Zhenfang Zhao begin their analysis by noting a set of stylised facts concerning high-frequency financial returns series, namely volatility clustering, high kurtosis, slow-decaying autocorrelation in squared returns, and so forth. In order to evaluate the capacity of volatility models to reproduce these facts, they apply both standard and robust measures of kurtosis and autocorrelation of squares to asset of first-order volatility models. Robust measures are particularly important in this exercise since many financial time series may be affected by outliers.

In their contribution, Dick van Dijk, Haris Munandar and Christian M. Hafner examine the impact of the introduction of the euro on non-euro currencies. In particular they document the existence of large structural breaks in the unconditional correlations among the US dollar exchange rates of the British pound, Norwegian krone, Swedish krona, Swiss franc, and the euro during the period 1994–2003, using the framework of dynamic conditional correlation models. The authors find that such breaks occurred both at the time the formal decision to proceed with the euro was made in December 1996 and at the time of the actual introduction of the euro in January 1999 and were generally substantially lower during the intervening period.

Doing justice to the memory of Professor Sir Clive Granger the econometrician and applied economist and to the memory of Clive Granger the man is a difficult task: the stature is too high, the contributions too great, the man so gentle. I like to feel, however, that this collection, put together with the greatest affection as well as the highest respect by Sir Clive’s friends and colleagues, would have pleased him.

There goes Clive Granger: when comes such another?

The Applied Economics journals: a personal reflection*

Sir Clive W. J. Granger†

It gives me very great pleasure to open this conference to celebrate the ruby anniversary – that is to say, the 40th anniversary – of the Applied Economics journals. I am especially pleased to do so not only because of my belief in the importance of applied economics as an area of research and investigation and my belief in the value that the Applied Economics journals have given to the world over the last four decades, but also because of my own very long association with the journals: I was one of the first managing editors of the journals, I have published a number of papers in the journals (the first in 1971 and the latest, to date, in 2001¹), and I continue to this day as an advisory editor.

The Applied Economics family of journals began in 1969 with the publication of volume 1 of Applied Economics. I believe that, initially, the editorial duties were shared by Ralph Day and Michael Craft and this continued until the end of volume 3 in 1971. For volume 4 in 1972, Michael Craft became the sole editor. In 1973, for volume 5, it was agreed that there should be a pair of new editors, one chosen from within the editorial board and one from outside. Thus in 1973, I (from the board) and Maurice Peston became joint editors. I was then joint managing editor for 6 years until 1979, with volume 7, when Maurice Peston started his long run as sole editor, eventually ending in 1997. At that point Mark Taylor took over and he continues to the present day. Thus, the flagship journal of the Applied series has now been running for 40 years and over this period there have been just five editors. Of these, Maurice Peston was clearly the longest serving, over a period of 24 years.

Over the last 40 years, Applied Economics has evolved from four slim issues a year in volume 3 (1971), totalling 309 pages and with 25 articles (including one by myself, it so happens: Sowter et al., 1971), up to the present, with over 2500 pages in volume 39 in 2007. A single issue in volume 39 contained 390 pages and 32 articles, which is slightly more that the whole of volume 3. More recently we have seen the launching of the highly successful Applied Financial Economics, and Applied Economics Letters, dedicated to shorter papers, which I find very convenient; both of these journals are in the spirit of the flagship journal and provide an important service to the profession.

Over the years and around the world I have met many academics who told me that they published their first paper in one of the Applied Economics journals, and this has generated a lot of good will for the journal. (I understand that the present editor belongs to this group!).²

How the area covered has grown over the years! In the early days, around 1970, Applied Economics was essentially a British journal and I think that in those days most British universities did not have a Department of Applied Economics. Everyone had a Department of Economics, some had a Department of Industrial Economics or of Commerce, which would also cover finance and accounting. How the world has changed over the past 40 years!

In the early years, the majority of the articles in the journal came from the US and the UK. As a point of comparison, I looked over the contributors to the first nine issues of volume 40, covering the first half of 2008. This included over 100 papers in 1186 pages (it should be noted that the first issue consisted of a special issue, concerning the economics of gambling). On my count these issues contained contributions from authors in 22 countries! A list, in no special order, with the number of papers involved is as follows:³

• US – 31

• UK – 13

• Canada – 2

• Greece – 3

• Italy – 3

• Germany – 4

• Portugal – 2

• Spain – 5

• Norway – 1

• France – 4

• Belgium – 3

• Turkey – 3

• Netherlands – 4

• Taiwan – 13

• Hong Kong – 3

• Malaysia – 1

• South Korea – 1

• Philippines – 1

• Singapore – 2

• Japan – 3

• Fiji – 2

• Australia – 5

In summary, by region:

• North America – 33

• Europe – 45

• Asia and rest of world – 31

Some conclusions and comments concerning this wide variety would be as follows. First, submissions are coming from all over the world and the journal is providing a great service to the profession worldwide. I have not carried out a similar analysis for Applied Financial Economics or Applied Economics Letters, but I know that this is also true of those journals. Second, if the special issue on gambling is subtracted, the UK figure drops to just 3, which seems to be rather low; the special issue has little effect on other numbers. Third, inside the front cover, 25 editors of various kinds for the journal are listed, with 12 based in the UK, nine from the US, two from Italy and one each from Spain and the Netherlands: spatially, the spread of the editors is rather different than that of the contributors.

The Applied journals have as their common denominator the requirement that all papers published have an applied aspect in some field of economics. How should a reader evaluate the quality of a paper in the general field of applied economics? My own view is that applied papers in these journals should be judged by four criteria. First, there is the importance and relevance of the topic. I think that it needs to be up to date and currently topical to be relevant: there are specialist journals dealing with economic history. Second, there is the quality and quantity of the data; I often observe the surprising fact that researchers know very little about the quality of their data! Third, there is the relevance and quality of the economic theory underlying the applied analysis; sometimes the theory quoted in highly applied work is of little relevance because it is based on unrealistic assumptions that are clearly violated in practice. Fourth and finally, the same applies to the quality and relevance of the econometric methods utilized.

In this context, a question that naturally arises is whether or not the author of an applied piece should try to present a single ‘best’ model of several appropriate models, so that the outcomes can be compared, discussed and combined; to my mind, it is clearly preferable to consider the results from several models.

It is also my belief that, in a really solid piece of applied work, the purpose of the work should be clearly presented together with a statement about which group of economic decision makers should most benefit from it. My own thinking on the fundamental question, what is the purpose of economics?, was influenced by a statement that I came across in the literature, saying that, economics is a decision science; I took this to mean that the purpose of economics should be to help decision makers make better decisions. Without going into the relative abilities of economic theory or econometrics to help achieve this purpose, it is quite clear that applied economics as an area is exactly on...

Table des matières

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• 1. Special issue in honour of Clive Granger
• 2. The Applied Economics journals: a personal reflection
• 3. Modelling the Phillips curve with unobserved components
• 4. Revisiting UK consumers’ expenditure: cointegration, breaks and robust forecasts
• 5. Combining forecasts – forty years later
• 6. Some variables are more worthy than others: new diffusion index evidence on the monitoring of key economic indicators
• 7. Does news on real Chinese GDP growth impact stock markets?
• 8. Stylized facts of return series, robust estimates and three popular models of volatility
• 9. The euro introduction and noneuro currencies
• Index