Quantitative approaches and modeling in marketing research

Journal of Modelling in Management

ISSN: 1746-5664

Publication date: 11 November 2014

Citation

Manrai, A. (2014), "Quantitative approaches and modeling in marketing research", Journal of Modelling in Management, Vol. 9 No. 3. https://doi.org/10.1108/JM2-07-2014-0063

Publisher

:

Emerald Group Publishing Limited


Quantitative approaches and modeling in marketing research

Article Type: Guest editorial From: Journal of Modelling in Management, Volume 9, Issue 3

Introduction

The idea of the special issue was conceived in 2012 when the editor-in-chief, Dr Luiz Moutinho invited me to serve the Journal of Modelling in Management (JM2) in the role of a Guest Editor for a special issue. What ensued was a call for papers in May 2012 and an intense activity of submissions, reviews and revisions. In about two years, the special issue was completed with acceptance of four excellent manuscripts. The same are discussed briefly in this editorial.

First paper

The first paper by Srinidhi and Manrai develops two analytic models of international airline demand. The paper analyzed macroeconomic factors to identify parameters as direct and indirect drivers of international air travel. It used a combination of cross-section and time series data to develop two analytical models. The zonal income function was found to perform better in terms of identifying demand drivers. By using smoothing functions on the demand series and forecasts until 2020, the paper provides airlines a road map to design future route identification strategies. The results and analysis presented in the paper suggest that airlines need to look at more than conventional variables for route contemplation and fare policies. This research will help airlines in route planning, purchasing decisions on aircraft and fuels, ticket pricing and other policy decisions. Additionally, this work will provide airlines a strong ground to develop code shares or partners which in-turn result in an improvised hub-spoke network; for example, while it may be viable for Air India to fly to Europe, it may not be viable to fly to Canada due to diminishing onward demand.

There are implications for policymakers as well. With a few international carriers considering cutting on operations or pushing for tie-ups, options dwindle for the consumers. Thus, slashing tax rates particularly the value-added tax, which is in the range of 20-30 per cent on airline turbine fuel (ATF) or revising import duties would make ATF cheaper, thus attracting more investment in terms of operations simultaneously reviving domestic carriers such as Kingfisher to fly on international routes. Although airport charges are considerable, ATF costs constitute more than 40 per cent of the total costs to airline companies; it would payoff for the government to allow airlines to directly import ATF whereby they would avoid paying sales tax varying between 4-32 per cent in different states. A better tax regime for aviation in general and ATF in particular will widen the economic benefits available from aviation thereby inducing a positive impact on economic growth and overall government revenue bases. The empirical section of the research paper developed for the Indian context is generalizable and will help the airlines in better route contemplation whilst formulating more important policy decisions.

Second paper

The second paper by Andrews and Ebbes examines the endogeneity problems in demand models. The endogeneity can bias the elasticities of the endogenous variable and subsequent optimization of the marketing mix. It is well known that, to remedy the endogeneity problem in linear regression models, the use of instrumental variables (IV) techniques with poor-quality instruments can produce very poor parameter estimates. The two main questions addressed in this paper are:

1. What are the properties of various IV estimators in logit-based demand models in which prices are potentially endogenous but instruments backed by theory are potentially weak and/or endogenous?

2. How should an analyst manage a potential endogeneity problem in such applications?

Using simulation methods, the paper investigated the effects of using poor-quality instruments to remedy endogeneity in logit-based demand models applied to finite-sample data sets. The results show that, even when the conditions for lack of parameter identification due to poor-quality instruments do not hold exactly, estimates of price elasticities can still be quite poor. The paper then investigated the relative performance of several nonlinear instrumental variables estimation procedures utilizing readily available instruments in finite samples. It is shown that endogeneity problems are exacerbated with increases in the number of brands, especially when poor-quality instruments are used. In addition, the number of stores is found to be important for likelihood ratio testing. The results of the simulation are shown to generalize to situations under Nash pricing in oligopolistic markets, to conditions in which cross-sectional preference heterogeneity exists and to nested logit and probit-based demand specifications as well. Based on the results of simulation, the paper suggests a procedure for managing a potential endogeneity problem in logit-based demand models.

Third paper

The third paper by Echchakoui provides justification for the importance given to sales force as a prominent component in the development of the supplier–customer relationship, which lacks attention in the extant literature. The author employed a dynamic exchange model between firm and salesperson as well as transaction cost economics (TCE) theory to explore the conditions affecting salesperson profitability in a relationship perspective. The application of this model reveals that customer value, salesperson cost, customer–salesperson relationship duration and firm’s margin in the absence of the salesperson have an impact on salesperson profitability. The study presents several implications for sales manager who wishes to focus on long-term relationship with customers. The proposed simplified categorization of the cases is an initial step, so that firms can appreciate that a return on investment following the recruitment of a salesperson is not guaranteed, but is dependent on certain parameters, such as customer value, salesperson cost and the duration of the salesperson–customer relationship. In this respect, research results specify the critical duration of the salesperson–customer relationship, which may be used as a basis for determining the probation period for a new salesperson. Like this probation period may vary from one salesperson to another, depending on the customer portfolio assigned to the salesperson, as well as the compensation paid.

The results of the study may also encourage managers to adopt different strategies when it comes to recruiting the sales force, depending on the nature of the customer portfolio. Furthermore, an effective relationship orientation requires the tailoring of services to the level of each potential customer. The success of such an action requires customized compensation to motivate salespersons to follow their firm’s strategy. It is argued that customized compensation can be used to encourage salespersons to further develop their skills and build upon their relationships with customers. It shows, the more a salesperson develops his/her relationships with customers, the higher his/her compensation.

Fourth paper

The final and fourth paper by Siddiqi focuses on several tests available in the statistical literature to assess normality of a given set of data (observations.) The author argues that not all these tests are suitable for all situations but each test has an exclusive area of application. The paper shows how these tests perform in normal and near normal situations. Using simulation experiments the, paper evaluated the tests for their power for various values of skewness, kurtosis and sample size. It showed that almost all these tests are indifferent for smaller values of skewness and kurtosis. Further, the power of accepting normality reduces with increasing sample size. These simulations identify tests, which perform comparatively better in almost all the situations while some other tests, which perform poorly. The author cautions that no verdict should be considered final. Setting the scenario, from practical point of view, and especially for non-technical researchers, it is quite wise to start with visual inspection of the data either through QQ plots, density plots or some other more intelligent graphical techniques, to have a sketch of the data. Such a sketch would be helpful not only to appraise the correctness of the diagnostics but also to locate the abnormality in the data, if it exists. It is recommended never to rely on the results of a single diagnostic test but apply multiple tests and to use tests that give better results in most situations, and furthermore, the results of the numerical diagnostics should be read in conjunction with the graphical sketch.

The guest editor greatly appreciates the opportunity, general guidelines, support and suggestions provided by the editor-in-chief, Dr Luiz Moutinho. The inspiration, comments and reviews provided by Dr Lalita Manrai were very helpful in preparing the special issue. The success of the special issue was made possible with the time, effort and diligence of several reviewers. The compilation of the special issue was also greatly helped by the timely and expert support of Wendy Weaver at the University of Glasgow and Chris Harris at Emerald Group Publishing. The invaluable support of the Department of Business Administration, College of Business and Economics, University of Delaware, by way of equipment use, mailing and time contributed a great deal to make this issue possible.

Ajay K. Manrai