In this paper we try to review few econometric journals to explore the structure and topics covered by the articles included in these journals. We try to shed some light on the data sources and types used in the researches, as well as the methodologies and models exploited to analyze the data. By skimming through the journals, we also try to discover the change in pattern of the articles composition from decades back till recent time. Eventually, we spot area of interest for future studying and further researching.
- Journal of Financial Econometrics
- Quantitative Marketing and Economics
- Econometric Reviews
Journals Scope and Review
Econometrica is one of the leading econometrics journals that aim at covering a wide range of economics topics (micro & macro). The journal started in 1932 with its first volume, reaching its 77th volume in 2009. Each volume has six issues that cover two months. Econometrica includes research studies that introduce new theory that adds up to the econometrics body of knowledge as well as studies that further investigate existing economic theories. The research included in the journal has both, the quantitative and qualitative component of the research. Almost all of the research topics start off by introducing the qualitative aspects of the research and then move towards the quantitative analysis via statistical models. Through our review of the journal, we noticed that the journal covers cross-sectional & time series analysis. There are many articles that have cross-sectional and time series analysis combined. (Econometric Society, 2009)
The aim of Econometrica is to bring the theory and experimental sides of studies and merge them together. Econometrica is not constricted to one topic or aspect; however, it covers a wide range of topics ranging from research practiced by a variety of researchers, different theoretical aspects of many issues, to productive studies constructed in econometrics. It is well known to act as a backbone to many of the research done and is trusted to be a very beneficial and supporting journal with sound history. It helps produce the best of the best in econometric research and development of various topics to maintain the highest standard in the submission of diverse studies related to econometrics.
Another leading journal of economics is Economica. It first began at around the year 1933, and it contains 76 volumes published yearly. In addition, each volume contains from one to four issues with quarterly publications. Economica provides researches in various aspects of economics in both experimental and theoretical studies constructed by various worldwide researchers with a great focus on quantitative analysis and econometric related topics. (London School of Economics, 2009)
A third journal that we came across was a fairly recent journal that was established in the year 1998: The Econometrics Journal. The purpose behind establishing this journal was to accommodate international publications related to econometrics, from all aspects, whether macro, micro, or finance-related econometrics. This journal ensured keeping the same standard that was previously shown in the past well known econometric-related journals. All papers that were to be published in The Econometrics Journal were thoroughly examined and very closely inspected to make sure that very high quality publications were only included. (Royal Economic Society, 2009)
The Econometrics Journal is very unique in the sense that it offers members of the Royal Economic Society, who initially established this journal, along with over three thousand users the chance to immediately have access to publications online, unlike other journals that delay online publications. (Royal Economic Society, 2009)
Furthermore, another article we found was the Journal of Financial Econometrics which consists of six volumes and each volume contains from one to four issues. This journal is aimed at becoming one of the top leading Journals amongst subject-related journals. It focuses on creating a link between econometrics and finance through both theoretical and experimental ways. The focal point in this journal is associating risk management and asset pricing with the field of econometrics in order to cover issues raised by the financial industry. (Journal of Financial Econometrics, 2009)
The fifth journal we in our review is the Quantitative Marketing and Economics Journal, which consists of seven volumes with each volume containing four issues. This journal is considered one of the recent journal since it is only dated back to the year 2003 where the first volume came out. The journal objective is to bond Marketing, Economics and Statistics research together to relate any similarities between them. This journal tries to explain marketing concepts in a quantitative matter because, as known, marketing is very broad in definition and can be linked to almost all concepts and areas that firms deal with. That is from consumers’ behaviors, consumers’ choices, preferences, to branding, pricing, promotion, positioning, etc. These concepts are used in this journal to create a link with both economics and statistics, hence the quantitative side of it. The QME journal uses both primary and secondary data to come up with the research conducted by the writers. (Marketing, 2004)
The final journal that we came across was the Journal of Econometric Reviews. This journal is very wide in the sense that it consists of twenty eight volumes with each volume having from one to six issues, the first volume starting in the year 1982. The Journal of Econometric Reviews aims at providing unlimited resources related to econometric matters. It offers readers top quality articles and publications on several themes in econometrics as well as it provides a leading editorial board who bring together world class experts and professionals to come up special editions of articles and papers for those who are interested. (Econometrics Reviews, 2008)
Data & Data-Sources of the Journals:
Through browsing the journals, we’ve came across many topics that involve substantial data gathering. The data ranged from publicly available data (GDP, interest rates etc…) to proprietary data that was collected through specialized information agents (daily sales data in a supermarket, number of customer foot-fall through the help of scanning devices etc…). Some examples of the data that we came across include:
- For a research that measures the implications of sales and consumer demand, data pertaining to sales and inventory level were gathered through Information Resource Inc. for the period of June 1991 to June 1993. The data was gathered by placing scanners in 9 supermarkets to collect information about the product, price, quantity purchased, customer info etc… (Hendel & Nevo, 2006)
- Data related to employee compensation for a study in France from 1993 to 2000 about the wage bargaining theory between the employee and employer. The data used are collected via the French National Statistical Institute INSEE. Furthermore, the research uses DADS (D�clarations Annuelles de Donn�es Sociales) as a data source that is used to derive labor costs at the firm level for all kinds of employee talent. (Cahuc, Vinay, & Robin, 2006)
- A research paper that studies the effect of communication on trust and cooperation gathers data by having live feedback from the participants. The sessions of the study were conducted in the University of California where the participants (24-36 per session) were seated in groups. The sessions also included the experimental instructors who administered the session for the data gathering. Examples of the data collected can be seen below in figure 1. (Charness, & Dufwenberg, 2006)
- A research about the labor mobility and the growth of the service sector gathers data on employment, wags, sector output and capital through the Bureau of Economic Analysis from 1968 – 2001 (Lee, & Wolpin, 2006). Samples of the data are shown in figures 2, 3 & 4 below:
Reviewing the journals, we have noticed a quite common style in structuring the articles and a major dependence on using hypothesis testing format. Depending on the depth and scope of the research, most articles are quantitative in nature trying to understand and relate actual occurrences of incidents to major econometric theories, whereby data is tested via models to prove significance and identify relationships that lead to better understanding of the dependent variable and pave the road for forecasting future occurrences.
The general steps in conducting econometric researches involve the following:
- Formulate a model
- Gather the data
- Estimate the model
- Subject the model to hypothesis testing
- Interpret the results
Initially, most articles start by reviewing some econometric literature prior to proceeding with the empirical work. Previous articles and theories are studied and discussed and analyzed with references to contemporary theories and hypothesis that might be the basis for the authors to carry on their research and build on others findings and conclusions.
In addition, there are few other articles with a qualitative base aimed at analyzing other researches and articles already published in a logical, scientific and rational approach. These articles tend to use logical methodologies to object or support other findings and hypothesis in contention. They can be in the form of pointing out conflicts or coherent parts of the economic literature to the issues and topics under research.
There is also a general trend among large articles to have a preliminary empirical and statistical analysis of the data in their possession. Some of the main methods used to conduct early stage checking include:
- Box plots,
- Checks for the presence of outliers,
- Summary statistics such as mean, standard deviations, coefficient of variation
- Correlation matrix etc…
Conducting a preliminary analysis usually guide the author to the model and methods to be used in researching his data. Early articles published during the 1970s and 1980s tend to rely on single equation methods model, where a single variable is used as a function of one or more explanatory variables. This method was criticized in some articles for possessing poor statistical properties or may not recover the effect desired. Whereas, more contemporary articles tend to use simultaneous equation methods that use variants of instrumental variables to make the estimates.
Furthermore, financial econometric models had significant developments in the last decade. With number of articles rising substantially than earlier times, due to the rapid growth in financial industry and increasing sophistication of financial methodologies. The toolkit of financial econometrics has grown in size and depth, including techniques like:
- Nonparametric estimation,
- Functional central limit theory,
- Nonlinear time-series models,
- Artificial neural networks,
- Markov Chain Monte Carlo methods.
- GARCH processes
- Normal Inverse Gaussian distribution etc…
For instance, an article named “A Bayesian methodology for simultaneously detecting and estimating regime change points and variable selection in multiple regression models for marketing research” in the Quantitative Marketing and Economics journal used a Bayesian change point multiple regression methodology. This is thought to enable the researchers to estimates the coefficients of the variables i.e. location of change points/regimes and their corresponding subset per regime, as well as the associated regimes’ regression parameters.
While in the Journal of Financial Econometrics, the authors of “Bias-Reduced Estimation of Long-Memory Stochastic Volatility” used a stochastic volatility model with potential nonstationarity in the volatility process to estimate the memory parameter in volatility.
Likewise, the maximum entropy method was used in “Optimal Portfolio Diversification Using the Maximum Entropy Principle.” of the Econometric Reviews Journal, in order to oppose the mean variance (MV) approach that often leads to portfolios being highly concentrated on a few assets and thus leads to poor out-of-sample performances.
Also, in a paper titled “Least Squares Model Averaging”, Econometrica, that considers the problem of selection of weights for averaging across least squares estimates obtained from a set of models, the Mallows model average (MMA) estimator was used to achieve the lowest possible squared error in a class of discrete model average estimators.
Time Series Methods
Most time series articles are common while analyzing economical and financial issues. Majority of papers referring to pricing models, consumer utility models and other macroeconomic cases are studied with data collected over time and hence time series methods were used.
However, recent volumes and issues on most econometric journals, especially in the last four years, have used a combination of statistical methods. Some used time series models across a range of variables and then used cross sectional methods to compare between them. Usually papers with panel data sources uses such a blend of methods.
Time series methods were quite common among early articles of the 1960s and 1970s as well as in most finance and accounting based articles. It is noticeable that the use of cross sectional methods dominated in the last decade articles. Most recent articles tried to use panel methods to combine both types and took advantage of the benefits of both to better explain the models in discussion.
We have also noticed that most macroeconomic quantitative researches were performed with time series methods. Probably it can be justified as most data was collected over time. Whilst, most customers linked articles and case studies of companies used cross sectional analysis methods. This is also justified due to the need that arised to compare data across a wide range of items and its effect on the model used.
One of the most active areas of research in econometrics that deserves a special consideration is financial econometrics. The demand for financial econometricians by investment banks and other financial institutions has never been greater, which rationalizes the rapid growth in the sophistication and complexity of models and tools used in such researches. This is evident in the evolution of the articles throughout the ‘Journal of Financial Econometrics’ and the increasing number of financial articles in the latest issues of most econometric and economic journals.
Many of the early 1970s authors have referred to the method of Least Squares in their articles as it was the main econometric tool used to analyze issues such as efficient markets, tests of the capital asset pricing model or arbitrage pricing theory, and stock returns forecasts. This is recently become inadequate to meet the new requirements of financial econometricians. The remarkable transformation in such techniques to more sophisticated ones can be attributed to three main reasons widely believed among authors of financial econometric articles:
- The set of breakthroughs in the quantitative modeling of financial markets.
- Contemporaneous set of breakthroughs in information technology.
- The rapid development in forecasting needs and methods.
The financial system has become more complex over time. The large expansion in the participants’ base and the large amounts involved has exerted pressure on the way financial researches were conducted and modeled. In the recent years econometricians continued to make serious and substantive contributions to the three pillars of finance: asset pricing, portfolio allocation, and risk management as identified by Francis X. Diebold in his publications. Such contributions have set the foundations to all modern quantitative financial analysis.
The second factor is related to the advances in information technology, including hardware, software, and data collection. The processing speed boost has provided a very effective computational medium for researches to use more algorithmic numerical and simulation methods. As described earlier, methods were such as: Bayesian econometrics using Markov chain Monte Carlo methods, bootstrap methods for inference, and model selection via intensive database search.
The third important factor of the observable hike in financial econometrics is the forecasting part of every academic research. Diebold mentions “forecasting is central to dynamic economics” which is highly relative to the high dynamics and robustness in modern finance. Almost all the articles related to financial quantitative analyses are aimed to initially understand and related the variables in contention and then plan to forecast and predict future occurrences.
Findings and Recommendations
Reviewing the journals has provided insights of economics and econometric topics and issues. Researchers are trying mainly to have a quantitative analysis of real world data in relationship with econometric theories and models in order to provide a systematic understanding of the data occurrences and its future forecasting.
In our opinion, there are a variety of topics and objectives in econometric articles that can be mainly categorized into three tranches:
- Topics that focus on analyzing public data to find a scientific explanations and relationships with other independent information. Financial econometric researches are a popular member of this category. Usually, financial information, such as, interest rates, stock prices, asset values, indices etc… are related to other factors like time, GDP, other economic etc… in order to find logical and statistical dependence and therefore, be able to analyze economic conditions, cycles and crisis.
- Other topics use a model or an econometric formula to extensively test its features and parameters using a wide range of data. Such articles relay on theoretical literature in order to explain how the model should function and on empirical testing to compare the results with the adequacy of the theory. For example, a model might work effectively in the absence of noise, but when noise is introduced the model is biased.
- Further topics pick an exiting research or hypothesis and try to proof its inefficient or try to add on to it as to make or more efficient and less vulnerable.
Our review of the journals has also revealed that researches with public reliable data are most common and has utmost publicity. This is due to the immediate reflection of these researches on the economic and professional environment and the costless data sourcing methods, with financial econometric and economic data analyses at the top of the hot topics list. While on the other hand, articles that introduce new models and equation and try to prove them, are the most difficult and less reached among researchers, yet, these topics can add the most to science and provide the maximum benefits to both academia and industry
We believe, that financial econometric has not been fully unexplored, and offer an opportunity to carry on studies and researches to grant better understanding of investments, finance and monetary activities.
We have explored few econometric and quantitative economic journals to review the topics and subjects covered by the articles included. A focus was on the data source that ranged from public information to explicit surveys and the data type such as time series, cross sectional and panel form. We have also noticed the extensive use of time series methods at earlier articles and journal issues, whereas, cross sectional and panel methods have gained high momentum in recent years. Moreover, we have paid special attention to financial econometric as the new development in the field of econometric and predicted abundant opportunities for further researches.
- Econometric Society, (2009). Econometrica, Journal of the Econometric Society. Retrieved from http://www.wiley.com/bw/journal.asp?ref=0012-9682&site;=1
- London School of Economics, Initials. (2009). Economica. Retrieved from http://www.wiley.com/bw/journal.asp?ref=0013-0427&site;=1
- Royal Economic Society, (2009). The Econometrics Journal. Retrieved from http://www.wiley.com/bw/journal.asp?ref=1368-4221&site;=1
- (2009). Journal of Financial Econometrics. Retrieved from http://www.oxfordjournals.org/our_journals/jfinec/about.html
- (2004, August 27). Marketing. Retrieved from http://www.springer.com/business/marketing/journal/11129?detailsPage=aimsAndScopes
- (2008). Econometrics Reviews. Retrieved from http://www.tandf.co.uk/journals/titles/07474938.asp
- Hendel, I., & Nevo, A. (2006). Measuring The Implications of Sales And Consumer Inventory Behavior. Econometrica, 74(6), 1637-1673.
- Cahuc P. Vinay F. & Robin J. (2006). Wage bargaining with on-The Job Search:Theory and Evidence . Econometrica, 74(2), 323-364.
- Charness, G., & Dufwenberg, M. (2006). Promises and Partnerships. Econometrica, 74(6), 1579-1601.
- Lee, D., & Wolpin, K. (2006). Intersectoral Labor Mobility and the Growth of the Service Sector. Econometrica, 74(1), 1-46.
- “Econometrics: Retrospect and Prospect”, Francis X. Diebold, 2001.