Faculty & Research

Research Productive

Show result

Search Query :
Area :
Search Query :
2757 items in total found

Working Papers | 2016

Nonlinear 0-1 knapsack problem with capacity selection

Sachin Jayaswal

We study a nonlinear 0-1 knapsack problem with capacity selection decision, as it arises as a part of facility location/service system design problems with congestion. The capacity selection decision gives rise to a non-convex objective function. We present two cutting plane based solution approaches: one based on Generalized Benders decomposition based, and the other based on a reformulation of the problem using additional auxiliary variables, followed by outer linearization of a resulting simple concave func-
tion in the constraint.

Read More

Working Papers | 2016

Learning Decision Models with Multinomial Logit Model through Pair-wise Preferences

Manish Aggarwal

Our goal is to study the behavioral process of a decision maker (DM) that leads to his choice.
To this end, we combine the established models of discrete choice with the recent algorithmic
advances in the emerging field of preference learning. Our proposed model takes the learning
information in form of the exemplary preference information, as revealed by a DM, and returns
the DM's choice probability. To accomplish our learning objective, we resort to the probabilistic
models of discrete choice and make use of the maximum likelihood inference. First experimental
results on suitable preference data suggest that our approach is not only intuitively appealing and
interesting from an interpretation point of view but also competitive to state-of-the-art preference
learning methods in terms of the prediction accuracy.

Read More

Working Papers | 2016

Interactive Discrete Choice Models

Manish Aggarwal

The interaction among the criteria plays an integral part in the human decision making
process. We propose new logit models of discrete choice to give the choice probability, taking into consideration the interaction among the attributes. We further extend the proposed models to also consider the complex human attitudinal character along with the interaction among the attributes. Concretely, we combine the established econometric models with Choquet integral, attitudinal Choquet integral, Choquet integral and generalized means, and generalized attitudinal Choquet integral. The proposed models of discrete choice can be seen as providing an overarching framework with the existing models as its special cases. The scope of the existing models are significantly expanded through the generation of a very wide range of choice probabilities arising in accordance with the adjustable attitudinal parameter(s).

Read More

Working Papers | 2016

On the Class of Attitudinal Discrete Choice Models

Manish Aggarwal

The complex human attitudinal character plays an important role in the real world decision
making. To this end, we present a family of extended probabilistic discrete choice models. The
attitude-based variants of multinomial logit, probit, nested, and mixed multinomial models are
presented. The proposed models are further empowered through their generalization leading to a
host of exponential attitudinal discrete choice models. It is shown that the existing models are
the special cases of the proposed models that allow to generate a very wide range of choice
probabilities in accordance with the adjustable parameter(s). The usefulness of the proposed
models in modelling a decision-maker's decision model is shown in a sequel paper.

Read More

Working Papers | 2016

Information Utility based Decision Support Framework

Manish Aggarwal

We introduce a novel entropy framework
for the computation of utility on the basis of an agent's subjective evaluation of the granularized information source values. A concept of evaluating agent as an information gain function of this entropy framework is proposed, which takes as its arguments both an information source value and the agent's evaluation of the same. A method to determine the agent's perceived utility values is also developed. Based on these values, several
new utility measures are designed for the valuation of the information source values, perceived utilities, and the evaluating agent.

Read More

Working Papers | 2016

Soft Information Set and its Application in Multi Criteria Decision Making

Manish Aggarwal

An information source value is perceived
differently by different agents. In this paper, we present a new knowledge representation structure, termed as soft information set (SIS), to provide a parameterized representation of the information values, as perceived by an agent. The properties of SIS are investigated and the
notion of relations in SIS is devised. SIS has potential to facilitate multi-criteria decision making (MCDM) under uncertainty, involving different information source values and agents. Its usefulness as a uncertainty representation
tool in MCDM is illustrated through case-studies.

Read More

Working Papers | 2016

A Game Theoretic Approach to Community based Data Sharing in Mobile Ad hoc networks

Premm Raj H. and Kavitha Ranganathan

Government interventions on usage of free speech
for communication has been rising of late. The government of Iraq's ban on the Internet, ban of mobile communications in Hong Kong student protests highlight the same. Applications
like Firechat which use mobile ad hoc networks (MANETs) to enable off the grid communication between mobile users, have gained popularity in these regions. However, there have been limited studies on selfish user behavior in community
data sharing networks. We wish to study these data sharing communities using game theoretic principles and propose a normal form game. We model selfishness in community data sharing MANETs and define the rationality for selfishness
in these networks. We also look at the impact of altruism in community data sharing MANETs and address the issue of minimum number of altruistic users needed to sustain the MANET. We validate the novel model using exhaustive
simulations and empirically derive important observations.

Read More

Working Papers | 2016

Web Content Analysis of Online Grocery Shopping Web Sites in India

Arindam Banerjee and Tanushri Banerjee

In this paper the authors evaluate online grocery shopping web sites catering to customers primarily in India. The process of evaluation has been carried out in 3 parts; by comparing the web content on their homepages, analysing customer reviews and also analysing their business performance as summarized on public web sites that use search optimization tools and analytical processes. This paper aims to study attributes from structured and unstructured data that lead to success of online grocery business in India. Results of the study will help identify the keywords that Indian consumers prefer to use while searching for information on online grocery websites. It will also identify consumer preferences from the customer review analysis. Additionally, it will identify the parameters from web site traffic metrics that drive per day revenue for the online retailer.

Read More

Working Papers | 2016

Learning of Utilitarian Decision Model through Preferences

Manish Aggarwal

Understanding and predicting the decision making behaviour of individuals is a
subject of interest for marketers, strategists, economists and the computer scientists
alike. We develop an aproach to learn a decision maker (DM)'s behavioral process
by combining recent possibilistic discrete choice models with the emerging machine
learning methods. The proposed approach considers the utility values derived by a
DM from each of the attribute values (information source values). We take the
training information in the form of the exemplary multi-attribute preferences, and
the decision model is specified in terms of two vectors that are unique to a DM. The
experimental results on a set of 10 benchmark datasets suggest that our approach is
both intuitively appealing and competitive to state-of-the-art methods in terms of the
prediction accuracy.

Read More

Working Papers | 2016

Classification Shifting: Impact of Firm Life Cycle

Neerav Nagar and Kaustav Sen

Purpose - This paper examines whether firms in the decline stage of life cycle manipulate core or operating income through misclassification of operating expenses as income-decreasing special items.
Design/methodology/approach - Our sample comprises of firms from an emerging market, India with data from 1996-2011. We use the methodology given in McVay (2006) and multiple regressions.
Findings - Managers of Indian firms also engage in classification shifting, primary incentive being desire to avoid reporting of operating losses. Further, the use of classification shifting is dependent upon the stage of life cycle in which firm is in. Specifically, firms in the decline stage of life cycle are more likely to use classification shifting to avoid reporting of operating losses.
Practical implications - The paper sheds light on a critical phase of the firm life cycle-decline, which increases the possibility of use of classification shifting-an earnings management technique which auditors, investors and regulators find tough to detect.
Originality/value - We extend the literature on classification shifting, and present first evidence that such shifting is more likely to take place during the decline phase of firm life cycle.

Read More
IIMA