Full text available at: http://dx.doi.org/10.

1561/1700000080

Methods for Causal

Inference in Marketing
 Full text available at: http://dx.doi.org/10.1561/1700000080

Other titles in Foundations and Trends® in Marketing

Aesthetics in Marketing
Henrik Hagtvedt
ISBN: 978-1-63828-286-0

Artificial Intelligence in Marketing and Consumer Behavior Research

TaeWoo Kim, Umair Usman, Aaron Garvey and Adam Duhachek
ISBN: 978-1-63828-266-2

Leveraging Online Search Data as a Source of Marketing Insights

Rex Yuxing Du and Tsung-Yiou Hsieh
ISBN: 978-1-63828-262-4

The Rise of Mobile Marketing: A Decade of Research in Review

Zeynep Aydin-Gokgoz, M. Berk Ataman and Gerrit van Bruggen
ISBN: 978-1-63828-098-9

Marketing and Firm Value

Shuba Srinivasan and Dominique M. Hanssens
ISBN: 978-1-63828-044-6
Full text available at: http://dx.doi.org/10.1561/1700000080

Methods for Causal Inference in

Marketing

Zezhen (Dawn) He
University of Rochester
[email protected]
Vithala R. Rao
Cornell University
[email protected]

Boston — Delft
 Full text available at: http://dx.doi.org/10.1561/1700000080

Foundations and Trends® in Marketing

Published, sold and distributed by:

now Publishers Inc.
PO Box 1024
Hanover, MA 02339
United States
Tel. +1-781-985-4510
www.nowpublishers.com
[email protected]
Outside North America:
now Publishers Inc.
PO Box 179
2600 AD Delft
The Netherlands
Tel. +31-6-51115274
The preferred citation for this publication is
Z. He and V. R. Rao. Methods for Causal Inference in Marketing. Foundations and
Trends® in Marketing, vol. 18, no. 3–4, pp. 176–309, 2024.
ISBN: 978-1-63828-343-0
© 2024 Z. He and V. R. Rao

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system,
or transmitted in any form or by any means, mechanical, photocopying, recording or otherwise,
without prior written permission of the publishers.
Photocopying. In the USA: This journal is registered at the Copyright Clearance Center, Inc., 222
Rosewood Drive, Danvers, MA 01923. Authorization to photocopy items for internal or personal
use, or the internal or personal use of specific clients, is granted by now Publishers Inc for users
registered with the Copyright Clearance Center (CCC). The ‘services’ for users can be found on
the internet at: www.copyright.com
For those organizations that have been granted a photocopy license, a separate system of payment
has been arranged. Authorization does not extend to other kinds of copying, such as that for
general distribution, for advertising or promotional purposes, for creating new collective works,
or for resale. In the rest of the world: Permission to photocopy must be obtained from the
copyright owner. Please apply to now Publishers Inc., PO Box 1024, Hanover, MA 02339, USA;
Tel. +1 781 871 0245; www.nowpublishers.com; [email protected]
now Publishers Inc. has an exclusive license to publish this material worldwide. Permission
to use this content must be obtained from the copyright license holder. Please apply to now
Publishers, PO Box 179, 2600 AD Delft, The Netherlands, www.nowpublishers.com; e-mail:
[email protected]
 Full text available at: http://dx.doi.org/10.1561/1700000080

Foundations and Trends® in Marketing

Volume 18, Issue 3–4, 2024
Editorial Board
Editor-in-Chief
Jehoshua Eliashberg
University of Pennsylvania

Editors
Dawn Iacobucci
Vanderbilt University
Leonard Lee
National University of Singapore
Sharon Ng
National University of Singapore
Koen Pauwels
Northeastern University
Stefano Puntoni
University of Pennsylvania
William Rand
North Carolina State University
Bernd Schmitt
Columbia University
Gerrit van Bruggen
Erasmus University
Hema Yoganarasimhan
University of Washington
Juanjuan Zhang
Massachusetts Institute of Technology
 Full text available at: http://dx.doi.org/10.1561/1700000080

Editorial Scope
Foundations and Trends® in Marketing publishes survey and tutorial articles
in the following topics:

• B2B Marketing • Marketing Decisions Models

• Bayesian Models • Market Forecasting
• Behavioral Decision Making • Marketing Information Systems
• Branding and Brand Equity • Market Response Models
• Channel Management • Market Segmentation
• Choice Modeling • Market Share Analysis
• Comparative Market Structure • Multi-channel Marketing
• Competitive Marketing Strat- • New Product Diffusion
egy
• Pricing Models
• Conjoint Analysis
• Product Development
• Customer Equity
• Product Innovation
• Customer Relationship Manage-
ment • Sales Forecasting

• Game Theoretic Models • Sales Force Management

• Group Choice and Negotiation • Sales Promotion

• Discrete Choice Models • Services Marketing

• Individual Decision Making • Stochastic Model

Information for Librarians

Foundations and Trends® in Marketing, 2024, Volume 18, 4 issues. ISSN
paper version 1555-0753. ISSN online version 1555-0761. Also available
as a combined paper and online subscription.
 Full text available at: http://dx.doi.org/10.1561/1700000080

Contents

1 Introduction 3

2 Origins and Developments in Causal Inference Methods 9

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Randomized Experiments . . . . . . . . . . . . . . . . . . 9
2.3 Potential Outcomes . . . . . . . . . . . . . . . . . . . . . 10
2.4 Illustrations of Marketing Experiments . . . . . . . . . . . 11
2.5 Treatment Effects . . . . . . . . . . . . . . . . . . . . . . 13
2.6 A Taxonomy of Data Types . . . . . . . . . . . . . . . . . 14
2.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

3 Data and Methods of Causal Inference 15

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.2 Directed Acyclic Graphs (DAGs) . . . . . . . . . . . . . . 16
3.3 Types of Data . . . . . . . . . . . . . . . . . . . . . . . . 17
3.4 Methods for Estimating Causal Effects . . . . . . . . . . . 18
3.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

4 Recently Developed CI Methods 37

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.2 Factor Model or Generalized Synthetic Control . . . . . . . 38
4.3 Augmented DID . . . . . . . . . . . . . . . . . . . . . . . 39
 Full text available at: http://dx.doi.org/10.1561/1700000080

4.4 Forward DID . . . . . . . . . . . . . . . . . . . . . . . . . 40

4.5 Bayesian Synthetic Control Methods . . . . . . . . . . . . 41
4.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

5 The Role of Machine Learning Methods in Causal

Inference 44
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 44
5.2 Selected Developments and Applications . . . . . . . . . . 45

6 Applications of Causal Inference Methods in

Marketing 50
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 50
6.2 Applications of the Differences-In-Differences
Regression Method . . . . . . . . . . . . . . . . . . . . . 51
6.3 Applications of the Instrumental Variable
Regression Method . . . . . . . . . . . . . . . . . . . . . 56
6.4 Applications of the Regression Discontinuity Method . . . 62
6.5 Applications of the Synthetic Control Method . . . . . . . 69
6.6 Applications of the Propensity Scoring Method . . . . . . 76
6.7 Applications of Recent Methods . . . . . . . . . . . . . . 83
6.8 Applications Using Machine Learning Techniques . . . . . 88

7 Conclusions and Future Research Directions 95

Acknowledgments 98

Appendices 99

A Python Code for Generating Simulated Data 100

B Stata Code for Analysis of Data 108

C ADID, Alternative Methods for ATT Estimation, and

Double Machine Learning 111

D Useful Resources 114

References 122
 Full text available at: http://dx.doi.org/10.1561/1700000080

Methods for Causal Inference in

Marketing
Zezhen (Dawn) He1 and Vithala R. Rao2
1 Simon Business School, University of Rochester, USA;
[email protected]
2 Johnson Graduate School of Management, Cornell University, USA;

[email protected]

ABSTRACT
Establishing causal relationships between the marketing
variables under the control of a firm and outcome measures
such as sales and profits is essential for the successful op-
eration of a business. The goal is generally hindered by a
lack of suitable experimental data owing to the costs and
feasibility of conducting randomized experiments. Accord-
ingly, researchers have employed observational and quasi-
experimental data for causal inference. The evolutionary
trajectory of causal inference is closely intertwined with
advancements in business and technology, particularly as
we enter the digital era characterized by big data and multi-
channel marketing.
Against this background, this monograph is a systematic
review of recent developments in causal inference methods
and their applications within the marketing field. For each
causal inference method, five recently published academic
papers in marketing research that employ these methods
are discussed.

Supplementary Material is available at http://dx.doi.org/10.1561/1700000080_supp

Zezhen (Dawn) He and Vithala R. Rao (2024), “Methods for Causal Inference in
Marketing”, Foundations and Trends® in Marketing: Vol. 18, No. 3–4, pp 176–309.
DOI: 10.1561/1700000080.
©2024 Z. He and V. R. Rao
 Full text available at: http://dx.doi.org/10.1561/1700000080

In addition, this monograph provides simplified code for

developing simulated data (using Python) and hypothetical
examples of data analysis (using Stata). This addition will
enable marketing researchers to practice several methods of
causal analysis.
Sections 1–5 elucidate the fundamental principles of causal
inference. Subsequent sections (beginning from Section 6)
delve into the details of a selection of papers that uti-
lize various methods. These encompass: (i) well-established
techniques, such as Differences-In-Differences, Instrumental
Variable, Regression Discontinuity, Synthetic Control, and
Propensity Score Methods, and (ii) emerging methodologies
of Factor Model and Augmented Differences-In-Differences,
Forward Differences-In-Differences, and Bayesian methods
for causal inference. Further, this monograph reviews how
machine learning methods enhance causal inference. This
monograph includes several important and useful references
not reviewed in the monograph. We expect this monograph
to serve as a useful resource both to current and future
researchers in marketing.
 Full text available at: http://dx.doi.org/10.1561/1700000080

1
Introduction

Marketing is a business function that relates the firm to its customers and
end-consumers. Undoubtedly, it is an essential and important function
of businesses and other organizations (see Kotler and Keller, 2012). On
the business side, the function involves activities such as product design,
sales forecasting, design of advertising strategy for a product and its
execution, and sales and distribution activities. The marketing function
is similar in other types of organizations, even though the terminologies
may differ.
Concurrent with the marketing developments in business, the aca-
demic field of marketing has blossomed over the last 80 years or so.
This discipline is thriving considering the high visibility of academic
marketing associations and academic journals. Three associations, the
American Marketing Association (AMA), the European Marketing
Academy (EMAC), and the Institute for Operations Research and Man-
agement Science (INFORMS) played a significant role in the discipline’s
growth. AMA’s premier journals, Journal of Marketing (JM) is in its
88th year of publication and the Journal of Marketing Research (JMR)
is in its 61st year of publication. The International Journal of Re-
search in Marketing (IJRM) of EMAC is in its 40th year of publication

3
 Full text available at: http://dx.doi.org/10.1561/1700000080

4 Introduction

Figure 1.1: Environmental factors affecting a firm.

Note: ∗ These include various social media and data collection firms.

while Marketing Science of INFORMS is in its 43rd year of publication.

Several other academic journals (e.g., the Journal of the Academy of
Marketing Science, the Journal of Consumer Research, and the Journal
of Retailing) cover the subject matter of marketing in several ways. In
a similar vein, universities have established doctoral degree programs in
marketing and are thriving well. The demand for faculty in marketing
is on the rise and is met by doctoral programs in marketing and other
subjects such as economics, psychology, and computer/information sci-
ence. All these point to the assertion that marketing is a viable field of
academic endeavor.
Figure 1.1 shows the external factors that affect a firm along with
the connections to intermediaries and consumers. It also shows the effect
of competitors. The intermediary firms include not only those firms
used for distributing the firm’s products but also other firms that collect
data on purchase and viewing behavior of end-consumers. The external
factors are categorized as Political, Economic, Social, Technological,
and Legal (or PESTLE for short).
While the field of marketing does not have a unified theory as such,
several important theories, paradigms, or frameworks have evolved over
the years, which are essential to both the practice of marketing and
research or theory building. The extant theories and paradigms include
the theory of buyer behavior (Howard and Sheth, 1969), Information
 Full text available at: http://dx.doi.org/10.1561/1700000080

Processing (Bettman, 1979), Hierarchy of effects of advertising (Lavidge

and Steiner, 1961), Customer lifetime value (Gupta and Lehmann, 2005;
Kumar, 2013) are used quite frequently. Further, the idea of “Marketing
Science” has taken hold; some of the research in this direction utilizes
game theory. There is an opportunity in marketing science to develop a
sound theoretical understanding of marketing issues.
On the applied side, several frameworks have appeared over the years.
These include the Four Ps (Product, Price, Place, and Promotion, also
called marketing-mix), STP (Segmentation, Targeting, and Positioning),
consumer choice process, and market response. These frameworks are
utilized to develop and implement a marketing plan for a firm’s product
(brand). Firms use several pieces of data developed from both internal
and external sources in the process of developing marketing plans.
External sources are typically focus groups, consumer surveys and
consumer panels, advertising test data, and occasionally experiments,
etc. The goal here is to come up with a plan that enables the firm
to reach a level of sales target for an existing product as well as a
new product (new brand). The research methods employed for this
purpose include several data analytic techniques (including multivariate
methods). Conceptually, the attempt here is to determine a market
response function for the product or brand in question that shows the
relationship between the sales outcome and variables that are in control
of the firm (i.e., the Four Ps). In general, the data sources within the
firm are not available to academic researchers.
In the early days of marketing research, the tendency was to re-
port summaries of data collected. Currently, the emphasis is on the
development of appropriate models and estimation using advanced and
appropriate methods.
Over the years, academic researchers have utilized data that they
collected on their own and also have utilized data sources that are
publicly available. Several new types of data have become available for
academic research as can be seen from Table 1.1. An example is the
data on online reviews of services like hotels (collected by companies
like Trip Advisor or Yelp). Another example is genetic data (Daviet
et al., 2022) that shed light on consumer behavior based on genetic
variants.
 Full text available at: http://dx.doi.org/10.1561/1700000080

6 Introduction

Table 1.1: Emerging types of data in marketing

A. Standard B. Unstructured C. Social

Marketing Data Data Network Data

A1. Survey data B1. Qualitative research C1. Social relationships data
A2. Experimental data∗ B2. Product/service reviews C2. Social games data
A3. Archival data B3. Videos, pictures C3. Postings to social media
A4. Panel data (choices B4. Consumer search data
and durations)
A5. Media ratings data B5. Data on physiological
measurements (e.g., eye
tracking)
A6. Sales and prices and B6. Neuroscience-related data
advertising data
B7. Genetic data

Note: ∗A special case of these data is conjoint analysis data (ratings or choices).

Academic researchers employ various methods in their research.

These methods are either drawn from statistics, econometrics, and psy-
chometrics or specially developed newer techniques. A few developments
in this area are the use of structural models, Hierarchical Bayesian (HB)
methods, and the application of newer methods drawn from AI (such
as deep learning). It is worth noting that the HB methods enable the
estimation of parameters at the individual unit (person, firm, or other)
level. One should underscore that the basic premise of sound research
is a viable theory translated into a model (preferably a mathematical
model). Whether explicit or not, academic research involves estimating
some model parameters, which are intended to measure the impact of
a predictor variable on the outcome variable (such as sales or choice).
If the researcher develops a system of equations to be estimated with
some econometric techniques, the coefficients of the model measure
such impacts. Depending on the theoretical basis for the research, the
estimate can be construed as a “causal effect.” As an example, con-
sider the Koyck model of dynamic effects of advertising specified as
St = α + β ∗ At + γ ∗ St−1 , where At is the advertising expenditures
in period t, and St is the sales in period t. If this theory is accepted,
then β is the causal effect of current advertising, and γ is the carry-over
effect.
Also, we may note that real-life experiments or randomized con-
trolled experiments out of the lab (Fisher, 1935) are expensive to carry
 Full text available at: http://dx.doi.org/10.1561/1700000080

out in marketing, even when researchers collaborate with industry. Even

though most marketing data are not based on designed experiments,
marketing researchers have employed data from quasi-experiments (also
called observational data1 ) for estimating causal effects. During the last
10 years or so, novel methods have emerged to estimate “causal effects.”
Researchers have also adopted methods developed for text analysis
as well; see Feder et al. (2022) for a comprehensive review of these
methods.
Two examples of studies that measure causal effects will be rele-
vant to mention here. One is the study by Wang et al. (2022) that
uses data from a natural experiment to determine the effect of the
Black Lives Matter movement on consumer responses (likes). They
employ the Differences-In-Differences (DID) method and estimate the
causal effect of the BLM movement on firms’ empirical strategy. The
authors exploit Blackout Tuesday as a natural experiment in which
BLM support occurred on Instagram (treated platform) but not on
Twitter (control platform) to perform a within-brand cross-platform
Differences-In-Differences (DID) analysis.
In a different context, Manchanda et al. (2015) measure the incre-
mental economic impact (expenditures) of customer communities (called
social dollars) in a multichannel retail environment. They measure this
by estimating a regression model. Because a well-defined control group
is missing, the authors test the effect of joining the online community
on economic activity using the Regression Discontinuity Analysis with
the time of joining the community as the threshold.
Against this background, the objective of this monograph is to
provide a comprehensive review of the methods for estimating causal
effects in marketing along with a review of the applications where these
methods have been applied. This monograph consists of seven sections.
In the next (and second) section, we offer an overview of the estimation
of causal effects attempting to synthesize the large literature in various
disciplines. The third section discusses various methods in brief; the

1
Although there are slight differences between observational data and quasi-
experimental data, we use them interchangeably in this monograph.
 Full text available at: http://dx.doi.org/10.1561/1700000080

8 Introduction

methods discussed are: Directed Acyclic Graphs,2 Analysis of Variance

and Covariance, Differences-In-Differences regression methods, regres-
sion with Instrumental Variables, Regression Discontinuity methods,
Synthetic Control Methods, as well as Sub-classification and Matching
methods (including Propensity Score Methods). In this section, we
attempt to provide the theoretical basis for each method along with
numerical examples and estimation codes. Section 4 describes four
emerging methods (i.e., Factor Model, Augmented DID, Forward DID,
and Bayesian Synthetic Control). The fifth section describes the role
and development of machine learning methods in causal estimation
with some examples. Section 6 reviews five applications of each of the
methods in marketing (i.e., Differences-In-Differences (DID), Instrumen-
tal Variable (IV), Regression Discontinuity (RD), Synthetic Control
Method (SCM), Propensity Score Method (PSM)), as well as emerging
methods and machine learning-related methods. Section 7 provides a
summary and a discussion of future directions in this rapidly growing
area.

2
This monograph will not cover the Directed Acyclic Graphs (DAGs) developed
by Pearl and his colleagues with an exclusive focus on econometrics-based methods.
See Pearl (2009a) for the DAG approaches. But we will briefly describe the debate
between Rubin and Pearl (Pearl, 2009b; Rubin, 1974).
Full text available at: http://dx.doi.org/10.1561/1700000080

Appendices
 Full text available at: http://dx.doi.org/10.1561/1700000080

A
Python Code for Generating Simulated Data1

Regression of Treatment Effects

Example Context: Assume a firm is interested in testing the impact of

a new television ad compared to its existing television ad, and the firm
will be airing the current ad in TV areas 1, . . . , 20 and airing the new
add in TV areas 21, . . . , 40. We define each area in terms of artificial
area codes i, where i = 1, . . . , 40, and denote each outcome measure as
Y (i, t) for t = 1, . . . , 10, where t is month.
The variables describing the TV areas are the average age of people
in the area, average income per household, the percentage of females in
the area, and the percentage of days in the period the brand was sold
on promotion.
The two outcome variables are the percentage of households buying
the brand during the period, and the percentage of buyers (households)
buying for the first time during the period.
The data is generated with the assumed treatment effects of 2 for
the percentage of households buying the brand during the period and
3 for the percentage of buyers (households) buying for the first time
during the period.

1
To enhance the replicability of the code, we have made the code available on
GitHub https://github.com/zhesimon/Methods-for-Causal-Inference-in-Marketing.

100
 Full text available at: http://dx.doi.org/10.1561/1700000080

101

import numpy as np
import pandas as pd
import random
from scipy.stats import norm
from sklearn.preprocessing import MinMaxScaler

np.random.seed(1)
TV_areas, n_periods = 40, 10
age_mean, age_std = 38, 5
income_household_mean, income_household_std = 70000, 10000
female_rate, female_std = 50, 10
pct_days_promo_mean, pct_days_promo_std = 50, 15

data = []
for i in range(TV_areas):
TV_area = i+1
age = int(np.random.normal(age_mean, age_std))
income = int(np.random.normal(income_household_mean, income_household_std))
female = round(np.random.normal(female_rate, female_std),2)
pct_days_promo = round(np.random.normal(pct_days_promo_mean,
pct_days_promo_std),2)

for j in range(n_periods):
TV_area_data = {
'period': j+1,
'TV_areas': TV_area,
'avg_age': age,
'avg_income': income,
'%female': female,
'pct_days_promo': pct_days_promo,
}
data.append(TV_area_data)
df = pd.DataFrame(data)

# outcome var, %households buying the brand

df['purchase_rate'] = (0.02 * df['avg_age'] + 0.5/10000 * df['avg_income'] + 0.1
* df['%female'] + 0.05 * df['pct_days_promo'] + np.random.uniform(-2, 2,
size = len(df)))

# outcome var, %buyers (households) buying for the first time

df['pct_buyer_1sttime'] = (0.01 * df['avg_age'] + 0.3/10000 * df['avg_income']
+ 0.05 * df['%female'] + 0.1 * df['pct_days_promo'] + np.random.uniform(-2, 2,
size = len(df)))

# trim values to between 0 and 100

df[['purchase_rate','pct_buyer_1sttime']] = df[['purchase_rate',
'pct_buyer_1sttime']].round(2).clip(0, 100)

# treatment group: TV_areas 21-40

df['treatment'] = [0 if i < TV_areas/2 * 10 else 1 for i in range(df.shape[0])]
df['purchase_rate_t'] = df['purchase_rate'] + 2 * df['treatment']
# treatment effect = 2
df['pct_buyer_1sttime_t'] = df['pct_buyer_1sttime'] + 3 * df['treatment']
 Full text available at: http://dx.doi.org/10.1561/1700000080

102 Python Code for Generating Simulated Data

# treatment effect = 3
df.to_csv('simulated_TV_areas.csv', index=False)

Nearest Neighbor Matching and Propensity Score Matching

Example Context: Here we used a similar dataset except that we gener-

ated the TV areas 21, . . . , 40 as close neighbors of TV areas 1, . . . , 20.
To do that, we duplicated the first 200 rows (thus, duplicated
TV_areas 1–20 that are control units) and added noise to the average
age of people living in the area, the average income per household, the
percentage of female in the area, and the percentage of days in the
period the brand was sold on promotion.
The covariates are the same with noise added, so the areas from the
duplicated rows are close neighbors of control areas.
# controls
new_df1 = df.iloc[:200, :6]

# new_df2 to be modified to close neighbors of controls

new_df2 = new_df1.copy()

new_df2['TV_areas'] = new_df2['TV_areas'] + 20

# error ~ N(0, 2) for age

new_df2['avg_age'] = df.groupby('TV_areas')['avg_age'].transform(lambda x:
x + np.random.normal(0, 2)).round(2)

# error ~ N(0, 1000) for income

new_df2['avg_income'] = df.groupby('TV_areas')['avg_income'].transform(lambda x:
x + np.random.normal(0, 1000)).round(2)

# error ~ N(0, 2) for percent female

new_df2['%female'] = df.groupby('TV_areas')['%female'].transform(lambda x: x +
np.random.normal(0, 2)).round(2)

# error ~ N(0, 2) for percent of days the brand was sold on promotion
new_df2['pct_days_promo'] = df.groupby('TV_areas')['pct_days_promo'].
transform(lambda x: x + np.random.normal(0, 2)).round(2)

df = pd.concat([new_df1, new_df2]).reset_index(drop=True)

# outcome var, %households buying the brand

df['purchase_rate'] = (0.02 * df['avg_age'] + 0.5/10000 * df['avg_income']
+ 0.1 * df['%female'] + 0.05 * df['pct_days_promo'] + np.random.uniform(-2, 2,
size = len(df)))

# outcome var, %buyers (households) buying for the first time

df['pct_buyer_1sttime'] = (0.01 * df['avg_age'] + 0.3/10000 * df['avg_income'] +
 Full text available at: http://dx.doi.org/10.1561/1700000080

103

0.05 * df['%female'] + 0.1 * df['pct_days_promo'] + np.random.uniform(-2, 2,

size = len(df)))

# trim values to between 0 and 100

df[['purchase_rate','pct_buyer_1sttime']] = df[['purchase_rate','pct_buyer_1sttime']]
.round(2).clip(0, 100)

# treatment for TV_areas 21-40

df['treatment'] = [0 if i < TV_areas/2 * 10 else 1 for i in range(df.shape[0])]
df['purchase_rate_t'] = df['purchase_rate'] + 2 * df['treatment']
# treatment effect = 2
df['pct_buyer_1sttime_t'] = df['pct_buyer_1sttime'] + 3 * df['treatment']
# treatment effect = 3
df.to_csv('simulated_TV_areas_nnmatch.csv', index=False)

Instrumental Variable (IV) Method

Example Context: We used the IV method to estimate the causal effect

of advertising on sales. For this purpose, we generated one unobserved
variable that affects both advertising expenditure and sales at the same
time, so there is endogeneity if we directly regress sales on advertising
expenditure. The instrumental variable is set as advertising costs. This
variable is generated to affect advertising expenditure but not sales
directly. We also assumed a linear relationship between advertising costs
and advertising expenditure, and between advertising expenditure and
sales.
We generated data for 100 periods, for each period there are different
advertising costs, advertising expenditure, and sales. The treatment
effect of advertising expenditure on sales is set to be 2.5.
np.random.seed(42)
n = 100 # periods
pi0 = 30 # intercept1
pi1 = -2 # effect of advertising costs on advertising expenditure
beta0 = 35 # intercept2
beta1 = 2.5 # effect of advertising expenditure on sales
omit_exp = 2 # effect of omitted variable on advertising expenditure
omit_sales = 3 # effect of omitted variable on sales
omitted_variable = np.random.randint(1, 3, size=n)

# IV (Z): advertising costs

advertising_costs = np.random.randint(1, 11, size=n)

# X: advertising expenditure
advertising_expenditure = pi0 + pi1 * advertising_costs + omit_exp * omitted_variable
+ np.random.uniform(-2, 2, size=n)
 Full text available at: http://dx.doi.org/10.1561/1700000080

104 Python Code for Generating Simulated Data

# Y: sales
sales = beta0 + beta1 * advertising_expenditure + omit_sales * omitted_variable
+ np.random.uniform(-5, 5, size=n)

data = pd.DataFrame({
'Ad_Costs': advertising_costs,
'Ad_Expenditure': advertising_expenditure,
'Omitted_variable': omitted_variable,
'Sales': sales
})
data.to_csv('IV_Data.csv',index=False)

Regression Discontinuity Method

Example Context: Assume a firm owns an online platform where hotel

guests provide ratings on three variables (brand image, price, and
service) after staying at a hotel.
Based on the overall feedback on the three variables (i.e., the average
score of brand image of the hotel, the average perceived price level, and
the average score of service performance), the firm develops a rating of
each hotel listed on the platform.
Further, the firm assigns symbols to the hotels based on a threshold.
If the rating is below 3, the platform assigns a Symbol B, and if the
rating is equal or above 3, Symbol A is assigned. Relative to Symbol B,
the treatment effect of being assigned Symbol A is 8.
The firm is interested in estimating the impact of being assigned to
Symbol A compared to Symbol B on hotel sales based on 1000 hotels.
We define each hotel in terms of hotel identification code i, where
i = 1, . . . , 1000, and denote each Sales measure as Y (i).
random.seed(0)
num_samples = 1000
beta1 = 8 # effect of symbol A on sales
beta2 = 5 # effect of ratings on sales

data = []
for _ in range(num_samples):
brandimage = random.uniform(1, 5)
price = random.uniform(1, 5)
service = random.uniform(1, 5)
rating = brandimage * 0.3 + price * -0.2 + service * 0.4

data.append({
'Brandimage': brandimage,
'Price': price,
 Full text available at: http://dx.doi.org/10.1561/1700000080

105

'Service': service,
'Rating_': rating})
df = pd.DataFrame(data)

# scale ratings to between 0 and 5

scaler = MinMaxScaler(feature_range = (1, 5))
df['Rating'] = scaler.fit_transform(df[['Rating_']])

# if rating < 3, assign symbol B, else assign Symbol A

df['Symbol'] = df['Rating'].apply(lambda x: 'B' if x < 3 else 'A')

df['Sales'] = beta1 * (df['Symbol'] == 'A') + beta2 * df['Rating']

+ df.apply(lambda row: random.uniform(-2, 2), axis = 1)
df.to_csv('RD_ratings.csv', index = False)

Synthetic Control Method (and Differences-In-Differences)

Example Context: Assume that a country consists of 51 geographically

identified States and that these States have a sales tax for products sold
within the State. Assume further that for a specific product category,
one State reduced tax in period 100 while other States did not. We are
interested in estimating the effect of the tax reduction on the State’s
product category sales using data available for 200 units of time (e.g.,
weeks). We define each State in terms of index i, where i = 0, 1, . . . , 50,
where State 0 is the treated unit, and States 1–50 are control States.
We denote each outcome measure as Y (i, t) for t = 1, . . . , 200, where
t is period. Periods 1–100 are pre-treatment periods, the treatment
happened in period 100, and periods 101–200 are post-treatment periods.
We assumed that the treatment effect is 10.
Initially, we generated the treatment unit using only two States i.e.,
State 1 (weight 0.2) and State 2 (weight 0.8) and random noise. Later,
we performed the Synthetic Control Method based on all 50 States,
States 1–25, States 1–10, and States 1–5. The pre-treatment periods
in our simulation were set as periods 1–100, periods 51–100, periods
81–100, or periods 91–100.
After obtaining the synthetic control, we used DID to estimate the
treatment effect.
random.seed(40)

# weights used for synthetic control (correspond to the 50 control states)

betas = [0.2, 0.8] + [0] * 48
 Full text available at: http://dx.doi.org/10.1561/1700000080

106 Python Code for Generating Simulated Data

# creates variables beta1, ..., beta50 and assigns them values from betas
for i in range(1, 51):
globals()[f'beta{i}'] = betas[i - 1]
print(beta1,beta2)
print(betas)

# generates a dictionary containing 50 key-value pairs, keys are "mu1", "mu2", ...,
,→ "mu50", values are random integers in range (5, 15)
mu_gen = {}
for i in range(1, 51):
mu_gen[f"mu{i}"] = random.randint(5, 15)
print(mu_gen)

# generates a list containing values of all mu's

mus = list(mu_gen.values())
print(mus)

# creates 50 variables mu1, ... , mu50, corresponding to the values stored in

,→ dictionary mu_gen
for i in range(1, 51):
globals()[f"mu{i}"] = mu_gen[f"mu{i}"]
#print(mu1, mu2, mu3)

control_units = ['Y' + str(i) for i in range(1, 51)]

#for i, Y in enumerate (control_units):

# print(control_units[i],mus[i],betas[i])

n=200
treated_period=int(n/2)
random_state=100
data= pd.DataFrame(index=range(n))

# control unit Y[i] ~ N (mu[i], 5)

for i, Y in enumerate (control_units):
random_state += 1
data[control_units[i]]=norm.rvs(loc=mus[i], scale=5, size=data.shape[0],
random_state=random_state)

# error term for the synthetic control ~ N (0, 1)

random_state=random_state + 1
data['error']=norm.rvs(loc=0, scale=1, size = data.shape[0], random_state =
random_state)

# data_w holds the weighted control units Y[i]w, weighted by their corresponding beta
,→ values
data_w=pd.DataFrame(index = range(n))
for i, Y in enumerate (control_units):
data_w[control_units[i] + 'w'] = betas[i] * data[control_units[i]]
data_w['error'] = data['error']
data_w.to_csv('Data_w.csv',index=False)

# treated outcome in the absence of treatment: weighted sum plus noise

 Full text available at: http://dx.doi.org/10.1561/1700000080

107

data['Y0_treated'] = data_w[list(data_w.columns)].sum(axis = 1)

# treatment effect for the treated units (post-treatment periods 101-200), which
,→ equals 10 + eps ~Uniform (-0.1,+0.1)
data['eps_te'] = np.concatenate((np.zeros(treated_period,dtype = int),
np.random.uniform(-0.1,0.1,treated_period) + 10))

# treated outcome in the presence of treatment

data['Y0_te'] = data['Y0_treated'] + data['eps_te']
data['period'] = np.arange(1,n + 1)

# check that the weighted Ys are Y*beta

for i in range(data.shape[0]):
assert(data['Y1'][i] * betas[0] == data_w['Y1w'][i])
data.to_csv('Synthetic_Control_Data.csv',index=False)
 Full text available at: http://dx.doi.org/10.1561/1700000080

B
Stata Code for Analysis of Data

Regression of Treatment Effects

import␣delimited␣"/simulated_TV_areas.csv",␣clear
rename␣female␣pct_female
*Dependent var: purchase rate
reg␣purchase_rate_t␣avg_age␣avg_income␣pct_female␣pct_days_promo␣treatment
*Dependent var: percent of first-time buyer
reg␣pct_buyer_1sttime_t␣avg_age␣avg_income␣pct_female␣pct_days_promo␣treatment

Nearest-Neighbor Matching
import␣delimited␣"/simulated_TV_areas_nnmatch.csv",␣clear
rename␣female␣pct_female
*Dependent var: purchase rate
teffects␣nnmatch␣(purchase_rate_t␣avg_age␣avg_income␣pct_female
pct_days_promo)(treatment),␣nneighbor(1)
*Dependent var: percent of first-time buyer
teffects␣nnmatch␣(pct_buyer_1sttime_t␣avg_age␣avg_income␣pct_female
pct_days_promo)(treatment),␣nneighbor(1)

Propensity Score Matching

import␣delimited␣"/simulated_TV_areas_nnmatch.csv",␣clear
rename␣female␣pct_female
*Dependent var: purchase rate
teffects␣psmatch␣(purchase_rate_t)␣(treatment␣avg_age␣avg_income␣pct_female
pct_days_promo)
*Dependent var: percent of first-time buyer
teffects␣psmatch␣(pct_buyer_1sttime_t)␣(treatment␣avg_age␣avg_income
pct_female␣pct_days_promo)

108
 Full text available at: http://dx.doi.org/10.1561/1700000080

109

Instrumental Variable
import␣delimited␣"/IV_Data.csv",␣clear
*Manually run 2sls
reg␣ad_expenditure␣ad_costs
gen␣constructed_ad_expenditure␣=␣_b[_cons]␣+␣_b[ad_costs]␣∗␣ad_costs
reg␣sales␣constructed_ad_expenditure

*IV
ivregress␣2sls␣sales␣(ad_expenditure␣=␣ad_costs)
*Test of endogeneity: check whether ad_expenditure is endogenous
estat␣endog
*Check whether the instrument is weak
estat␣firststage

Regression Discontinuity Method

import␣delimited␣"/RD_ratings.csv",␣clear
*RD plot, Cut-off c = 3
rdplot␣sales␣rating,␣c(3)␣graph_options(title(RD␣Plot)␣xtitle("Rating")
ytitle("Sales"))
*Sharp RD estimates
rdrobust␣sales␣rating,␣c(3)␣all

Synthetic Control Method

import␣delimited␣"/Synthetic_Control_Data.csv",␣clear
reshape␣long␣y,␣i(period)␣j(State)
rename␣period␣Period
rename␣y␣Sales
save␣"/Synth_Panel.dta",␣replace

*Performing synthetic control method using all 50 control states and all pre-treatment periods
use␣"/Synth_Panel.dta",␣clear
tsset␣state␣Period
synth␣Sales␣Sales(1(1)100),␣trunit(0)␣trperiod(101)␣fig␣keep(100period_50state)
graph␣export␣"/Graph_100period_50state.png",␣replace

*Performing synthetic control method using partial (25) control states and all
pre-treatment periods
use␣"/Synth_Panel.dta",␣clear
keep␣if␣State<26
tsset␣state␣Period
synth␣Sales␣Sales(1(1)100),␣trunit(0)␣trperiod(101)␣fig␣keep(100period_25state)
graph␣export␣"/Graph_100period_25state.png",␣replace

*Performing synthetic control method using all 50 control states and 50 pre-treatment periods
use␣"/Synth_Panel.dta",␣clear
keep␣if␣Period>50
tsset␣State␣Period
synth␣Sales␣Sales(51(1)100),␣trunit(0)␣trperiod(101)␣fig␣keep(50period_50state)
graph␣export␣"/Graph_50period_50state.png",␣replace
 Full text available at: http://dx.doi.org/10.1561/1700000080

110 Stata Code for Analysis of Data

*Calculating treatment effects using DID (reshape data from wide format to long format)
use␣"/100period_50state.dta",␣clear
drop␣_Co_Number␣_W_Weight
reshape␣long␣_Y_,␣i(_time)␣j(State,␣string)
gen␣treatment␣=␣0
replace␣treatment␣=␣1␣if␣_time␣>=101␣&␣State␣==␣"treated"
encode␣State,␣generate(nState)
xtset␣nState
xtdidregress␣(_Y_)(treatment),␣group(nState)␣time(_time)

Differences-In-Differences
use␣"/Synth_Panel.dta",␣clear
*DID result using State2 as control (which has weight of beta 0.8 when generating the treatment
unit in our simulation)
keep␣if␣State␣==2␣|␣State␣==0
gen␣treatment␣=␣0
replace␣treatment␣=␣1␣if␣Period␣>=101␣&␣State␣==␣0
xtset␣State
xtdidregress␣(Sales)(treatment),␣group(State)␣time(Period)
 Full text available at: http://dx.doi.org/10.1561/1700000080

C
ADID, Alternative Methods for ATT Estimation,
and Double Machine Learning

In this appendix, we briefly summarize the models behind the ADID

and its alternatives (DID, SC, MSC, and HCW) for estimating the
average treatment effect in quasi-experimental studies, and some notes
about Double Machine Learning.

ADID and Alternative Methods for ATT Estimation

After introducing the notation, we will describe the specific methods

and compare them with ADID.
The problem is to estimate the average treatment effect on the
treated units (or ATT) using data on different units, some of which are
treated for a number of time periods. For simplicity, we will assume that
the first unit is treated at time T1 ; 1 < T1 < T and all other (N − 1)
units do not receive treatment.
Notation:
(1)
yit = Outcome measure after treatment for the i-th unit at time t;
i = 1, . . . , N ; and t = 1, . . . , T .
(0)
yit = Outcome measure before treatment for the i-th unit at time
t; i = 1, . . . , N ; and t = 1, . . . , T .

111
 Full text available at: http://dx.doi.org/10.1561/1700000080

112 ADID, Alternative Methods for ATT Estimation

ˆ 1t = y (1) − y (0) = Treatment effect for the first unit at time t, with
∆ 1t 1t
t > T1 + 1.
We can describe the observed data as: yit = dit yit 1 +(1−d )y 0 , where
it it
dit = 1 if the i-th unit receives treatment at time t and 0 otherwise.
Given the above assumption on timing and units treated, the ATT
estimator that averages ∆ ˆ it over the post-treatment period is ∆ ˆ1 =
1 PT ˆ
T2 T1 +1 ∆it where T2 = T − T1 is the number of post-treatment time
periods., where ∆ ˆ it = y (1) − y (0) . While the data are available for y (0)
it it it
(1)
but not for yit , we need to estimate this quantity. The methods differ
according to the model used for this estimation. Table C.1 below shows
the models and differences.
To summarize, the ADID method is more flexible than DID because
it controls for slope in addition to intercept differences in pre-treatment
periods. This also means that it only requires that the treated unit’s
trend is parallel to a slope-adjusted trend of the control unit in pre-
treatment periods. Compared to Synthetic Control Methods that require
no intercept and weights sum to one, ADID require equal weights, but
the weights can sum to any value.

Double Machine Learning

Traditional machine learning methods are good at predictions, but not

good at causal relationships because they can produce biased estimates
due to regularization and overfitting. Double machine learning (Cher-
nozhukov et al., 2018a) is a recently developed framework that combines
machine learning methods and causal inference in that it deals with
regularization bias through orthogonalization and deals with overfit-
ting bias through sample splitting. Double√machine learning enables
in obtaining approximately unbiased and N -consistent estimation
under high dimensional complex covariates and confounding variables.
Researchers will find the Python and R packages of DoubleML (Bach
et al., 2021, 2022) valuable in the implementation of the double/debiased
machine learning framework of Chernozhukov et al. (2018a).
 Table C.1: A summary of formulae for different methods

Method Model for y 1t Formula for ŷ 01t Formula for ATT Comments
0 1 0
PT
ADID yit = δ1 + δ2 y co,t + error ŷ1t,ADID = δ̂1 + δ̂2 yco,t T2 (y1t − y̆ˆ1t,ADID ) Compared HCW, if βj s are equal,
T1 +1
; t = 1, . . . , T 1 yields ADID
0 1 0
PT
DID yit = δ1 + y co,t + error ŷ1t,DID = δ̂1 + δ̂2 yco,t T2 (y1t − y̆ˆ1t,DID ) Compared HCW, if βj s are equal
T1 +1
; t = 1, . . . , T1 and positive, and sum to 1,
yields DID
0 1 0
PT
SC y1t = zt0 β + error; βSC ŷ1t,sc = zt0 β̂sc T2 (y1t − y̆ˆ1t,SC ) Compared HCW, if βj s are all
T1 +1
minimizes error sum of positive and sum to 1 with no
squares intercept, yields SC
0 0 1 0
PT
MSC Same as for SC with the ŷ1t,M SC = zt β̂M SC T2 (y1t − y̆ˆ1t,M SC ) Compared HCW, if βj s are all
T1 +1
conditions,
PN β1 = positive, yields MSC
0; βj = 1; βj ≥ 0.
j=2
0 1 0
PT
HCW y1t = zt0 β + error; t = ŷ1t,HCW = zt0 β̂OLS T2 (y1t − y̆ˆ1t,HCW ) This is the base case against
T1 +1
1, . . . , T 1; β̂OLS is the least which comparisons are made.
squares estimate of β
Full text available at: http://dx.doi.org/10.1561/1700000080

113
 Full text available at: http://dx.doi.org/10.1561/1700000080

D
Useful Resources

Anderson, S. J., R. Chandy, and B. Zia (2018). “Pathways to profits:

The impact of marketing vs. finance skills on business performance”.
Management Science. 64(12): 5559–5583.

Angrist, J. D. (1990). “Lifetime earnings and the Vietnam era draft

lottery: Evidence from social security administrative records”. The
American Economic Review: 313–336.

Arslan, H. A., N. Tereyağoğlu, and Ö. Yılmaz (2023). “Scoring a

touchdown with variable pricing: Evidence from a quasi-experiment
in the NFL ticket markets”. Management Science. 69(8): 4435–4456.

Ascarza, E. (2018). “Retention futility: Targeting high-risk customers

might be ineffective”. Journal of Marketing Research. 55(1): 80–98.

Atanasov, V. and B. Black (2021). “The trouble with instruments: The

need for pretreatment balance in shock-based instrumental variable
designs”. Management Science. 67(2): 1270–1302.

Athey, S. (2015). “Machine learning and causal inference for policy

evaluation”. In: Proceedings of the 21st ACM SIGKDD International
Conference on Knowledge Discovery and Data Mining. 5–6.

114
 Full text available at: http://dx.doi.org/10.1561/1700000080

115

Athey, S. and G. W. Imbens (2017). “The state of applied econometrics:

Causality and policy evaluation”. Journal of Economic Perspectives.
31(2): 3–32.
Berger, J., A. Humphreys, S. Ludwig, W. W. Moe, O. Netzer, and
D. A. Schweidel (2020). “Uniting the tribes: Using text for marketing
insight”. Journal of Marketing. 84(1): 1–25.
Berman, R. and A. Israeli (2022). “The value of descriptive analytics:
Evidence from online retailers”. Marketing Science. 41(6): 1074–
1096.
Bertsimas, D. and N. Kallus (2020). “From predictive to prescriptive
analytics”. Management Science. 66(3): 1025–1044.
Bommaraju, R. and S. Hohenberg (2018). “Self-selected sales incen-
tives: Evidence of their effectiveness, persistence, durability, and
underlying mechanisms”. Journal of Marketing. 82(5): 106–124.
Bronnenberg, B. J., J.-P. Dubé, and R. E. Sanders (2020). “Consumer
misinformation and the brand premium: A private label blind taste
test”. Marketing Science. 39(2): 382–406.
Burtch, G., Q. He, Y. Hong, and D. Lee (2022). “How do peer awards
motivate creative content? Experimental evidence from Reddit”.
Management Science. 68(5): 3488–3506.
Chen, Y., J.-Y. Lee, S. Sridhar, V. Mittal, K. McCallister, and A. G.
Singal (2020). “Improving cancer outreach effectiveness through
targeting and economic assessments: Insights from a randomized
field experiment”. Journal of Marketing. 84(3): 1–27.
Cui, Y. and A. M. Davis (2022). “Tax-induced inequalities in the
sharing economy”. Management Science. 68(10): 7202–7220.
DeepLearning.AI (n.d.). “Natural language processing resources”. url:
https://www.deeplearning.ai/resources/natural-language-processing/.
Dubé, J.-P., G. J. Hitsch, and P. E. Rossi (2018). “Income and wealth
effects on private-label demand: Evidence from the great recession”.
Marketing Science. 37(1): 22–53.
 Full text available at: http://dx.doi.org/10.1561/1700000080

116 Useful Resources

Ertekin, N., J. D. Shulman, and H. Chen (2019). “On the profitability

of stacked discounts: Identifying revenue and cost effects of discount
framing”. Marketing Science. 38(2): 317–342.

Frisch, R. and F. V. Waugh (1933). “Partial time regressions as com-

pared with individual trends”. Econometrica: Journal of the Econo-
metric Society: 387–401.

Gelman et al. (2009). “Statistical modeling, causal inference, and social

science”. Statistical Modeling Causal Inference and Social Science.
url: http://statmodeling.stat.columbia.edu/2009/07/05/disputes_
about/.

Gentzkow, M., B. Kelly, and M. Taddy (2019). “Text as data”. Journal

of Economic Literature. 57(3): 535–574.

Goli, A. and P. K. Chintagunta (2021). “What happens when a retailer

drops a product category? investigating the consequences of ending
tobacco sales”. Marketing Science. 40(6): 1169–1198.

Golmohammadi, A., T. Havakhor, D. K. Gauri, and J. Comprix (2021).

“Complaint publicization in social media”. Journal of Marketing.
85(6): 1–23.

Gordon, B. R., M. J. Lovett, B. Luo, and J. C. Reeder III (2023a).

“Disentangling the effects of ad tone on voter turnout and candidate
choice in presidential elections”. Management Science. 69(1): 220–
243.

Gordon, B. R., R. Moakler, and F. Zettelmeyer (2023b). “Close enough?

A large-scale exploration of non-experimental approaches to adver-
tising measurement”. Marketing Science. 42(4): 768–793.

Gordon, B. R., F. Zettelmeyer, N. Bhargava, and D. Chapsky (2019).

“A comparison of approaches to advertising measurement: Evidence
from big field experiments at Facebook”. Marketing Science. 38(2):
193–225.

Green, P. E. and D. S. Tull (1988). “Research for Marketing Decisions”.

Prentice-Hall.
 Full text available at: http://dx.doi.org/10.1561/1700000080

117

Gu, X. and P. K. Kannan (2021). “The dark side of mobile app adop-
tion: Examining the impact on customers’ multichannel purchase”.
Journal of Marketing Research. 58(2): 246–264.

Guo, T., S. Sriram, and P. Manchanda (2021). “The effect of infor-

mation disclosure on industry payments to physicians”. Journal of
Marketing Research. 58(1): 115–140.

Gupta, S., D. Hanssens, B. Hardie, W. Kahn, V. Kumar, N. Lin, N.

Ravishanker, and S. Sriram (2006). “Modeling customer lifetime
value”. Journal of Service Research. 9(2): 139–155.

Habel, J., S. Alavi, and K. Linsenmayer (2021). “Variable compensation

and salesperson health”. Journal of Marketing. 85(3): 130–149.

Hassan, M., J. Prabhu, R. Chandy, and O. Narasimhan (2023). “When

bulldozers loom: Informal property rights and marketing practice
innovation among emerging market microentrepreneurs”. Marketing
Science. 42(1): 137–165.

He, C., O. C. Ozturk, C. Gu, and P. K. Chintagunta (2023). “Consumer

tax credits for EVs: Some quasi-experimental evidence on consumer
demand, product substitution, and carbon emissions”. Management
Science.

He, C., O. C. Ozturk, C. Gu, and J. M. Silva-Risso (2021). “The end

of the express road for hybrid vehicles: Can governments’ green
product incentives backfire?” Marketing Science. 40(1): 80–100.

Hermosilla, M., F. Gutierrez-Navratil, and J. Prieto-Rodriguez (2018).

“Can emerging markets tilt global product design? Impacts of Chinese
colorism on Hollywood castings”. Marketing Science. 37(3): 356–381.

IBM (n.d.[a]). “What is computer vision?” url: https://www.ibm.com/

topics/computer-vision.

IBM (n.d.[b]). “What is natural language processing (NLP)?” url:

https://www.ibm.com/topics/natural-language-processing.
 Full text available at: http://dx.doi.org/10.1561/1700000080

118 Useful Resources

Iyengar, R., Y.-H. Park, and Q. Yu (2022). “The impact of subscription

programs on customer purchases”. Journal of Marketing Research.
59(6): 1101–1119.

Jalali, N. Y. and P. Papatla (2019). “Composing tweets to increase

retweets”. International Journal of Research in Marketing. 36(4):
647–668.

Janakiraman, R., J. H. Lim, and R. Rishika (2018). “The effect of a

data breach announcement on customer behavior: Evidence from
a multichannel retailer”. Journal of Marketing. 82(2): 85–105.

Kallus, N. (2023). “Treatment effect risk: Bounds and inference”. Man-

agement Science. 69(8): 4579–4590.

Kamakura, W. A. and J. A. Mazzon (2015). “Measuring the impact of

a conditional cash transfer program on consumption behavior with
propensity scoring”. Customer Needs and Solutions. 2: 302–316.

Kim, T. and D. KC (2020). “Can Viagra advertising make more babies?

Direct-to-consumer advertising on public health outcomes”. Journal
of Marketing Research. 57(4): 599–616.

Krasnikov, A. and S. Jayachandran (2022). “Building brand assets:

The role of trademark rights”. Journal of Marketing Research. 59(5):
1059–1082.

Leeflang, P. S., D. R. Wittink, M. Wedel, and P. A. Naert (2013).

Building Models for Marketing Decisions. Vol. 9. Springer Science
& Business Media.

Li, K. T. and V. Shankar (2023). “A two-step synthetic control approach

for estimating causal effects of marketing events”. Management Sci-
ence. DOI: 10.1287/mnsc.2023.4878.

Lilien, G. L., P. Kotler, and K. S. Moorthy (1992). Marketing Models.

Vol. 803. Englewood Cliffs, NJ: Prentice-Hall.

Lim, J. H., R. Rishika, R. Janakiraman, and P. Kannan (2020). “Com-

petitive effects of front-of-package nutrition labeling adoption on
 Full text available at: http://dx.doi.org/10.1561/1700000080

119

nutritional quality: Evidence from facts up front-style labels”. Jour-

nal of Marketing. 84(6): 3–21.

Lovell, M. C. (1963). “Seasonal adjustment of economic time series and

multiple regression analysis”. Journal of the American Statistical
Association. 58(304): 993–1010.

Lu, S., K. Rajavi, and I. Dinner (2021). “The effect of over-the-top me-
dia services on piracy search: Evidence from a natural experiment”.
Marketing Science. 40(3): 548–568.

Luangrath, A. W., Y. Xu, and T. Wang (2023). “Paralanguage clas-

sifier (PARA): An algorithm for automatic coding of paralinguistic
nonverbal parts of speech in text”. Journal of Marketing Research.
60(2): 388–408.

Mitchell, R. (2018). Web Scraping with Python: Collecting More Data

from the Modern Web. O’Reilly Media, Inc.

Narayan, V. and S. Kankanhalli (2021). “The economic and social

impacts of migration on brand expenditure: Evidence from rural
india”. Journal of Marketing. 85(6): 63–82.

Neyman, J. (1979). “C (α) tests and their use”. Sankhyā: The Indian
Journal of Statistics, Series A: 1–21.

Neyman, J. and K. Iwaszkiewicz (1935). “Statistical problems in agri-

cultural experimentation”. Supplement to the Journal of the Royal
Statistical Society. 2(2): 107–180.

Palazzolo, M. and A. Pattabhiramaiah (2021). “The minimum wage and

consumer nutrition”. Journal of Marketing Research. 58(5): 845–869.

Park, H. M. and J. Pancras (2022). “Social and spatiotemporal impacts

of casino jackpot events”. Marketing Science. 41(3): 575–592.

Pattabhiramaiah, A., S. Sriram, and P. Manchanda (2019). “Paywalls:

Monetizing online content”. Journal of Marketing. 83(2): 19–36.
 Full text available at: http://dx.doi.org/10.1561/1700000080

120 Useful Resources

Pearl, J. (2018). “Theoretical impediments to machine learning with

seven sparks from the causal revolution”. arXiv preprint arXiv:1801.
04016.
Pearl, J. (2019). “The seven tools of causal inference, with reflections
on machine learning”. Communications of the ACM. 62(3): 54–60.
Rafieian, O. and H. Yoganarasimhan (2022). “Variety effects in mobile
advertising”. Journal of Marketing Research. 59(4): 718–738.
Raghu, B., A. Michael, R. Hall Zachary, T. Seshadri, and K. Lam Son
(2018). “The impact of mergers and acquisitions on the sales force”.
Journal of Marketing Research. 55(2): 254–264.
Ramani, N. and R. Srinivasan (2019). “Effects of liberalization on in-
cumbent firms’ marketing-mix responses and performance: Evidence
from a quasi-experiment”. Journal of Marketing. 83(5): 97–114.
Rao, R. S., M. Viswanathan, G. John, and S. Kishore (2021). “Do
activity-based incentive plans work? Evidence from a large-scale
field intervention”. Journal of Marketing Research. 58(4): 686–704.
Rao, V. R. (2006). Analysis for Strategic Marketing. India: Pearson.
Reza, S., H. Ho, R. Ling, and H. Shi (2021). “Experience effect in
the impact of free trial promotions”. Management Science. 67(3):
1648–1669.
Rossi, P. E. (2018). “Causal inference in marketing applications”. In:
Handbook of Marketing Analytics: Methods and Applications in Mar-
keting Management, Public Policy, and Litigation Support. Ed. by
N. Mizik and D. M. Hanssens. Cheltenham, UK: Edward Elgar
Publishing. Ch. 6.
Sakaguchi, H., N. Stewart, J. Gathergood, P. Adams, B. Guttman-
Kenney, L. Hayes, and S. Hunt (2022). “Default effects of credit card
minimum payments”. Journal of Marketing Research. 59(4): 775–796.
Schmitz, C., M. Friess, S. Alavi, and J. Habel (2020). “Understanding
the impact of relationship disruptions”. Journal of Marketing. 84(1):
66–87.
 Full text available at: http://dx.doi.org/10.1561/1700000080

121

Shi, Z., X. Liu, and K. Srinivasan (2022). “Hype news diffusion and
risk of misinformation: the Oz effect in health care”. Journal of
Marketing Research. 59(2): 327–352.

Shukla, A. D., G. Gao, and R. Agarwal (2021). “How digital word-

of-mouth affects consumer decision making: evidence from doctor
appointment booking”. Management Science. 67(3): 1546–1568.

Simonov, A. and S. Hill (2021). “Competitive advertising on brand

search: Traffic stealing and click quality”. Marketing Science. 40(5):
923–945.

Simonov, A., C. Nosko, and J. M. Rao (2018). “Competition and crowd-

out for brand keywords in sponsored search”. Marketing Science.
37(2): 200–215.

Viswanathan, M., X. Li, G. John, and O. Narasimhan (2018). “Is cash

king for sales compensation plans? Evidence from a large-scale field
intervention”. Journal of Marketing Research. 55(3): 368–381.

Wang, G., J. Li, and W. J. Hopp (2022) “An instrumental variable

forest approach for detecting heterogeneous treatment effects in
observational studies”. Management Science. 68(5): 3399–3418.

Wang, Y. and A. Chaudhry (2018). “When and how managers’ re-

sponses to online reviews affect subsequent reviews”. Journal of
Marketing Research. 55(2): 163–177.

Xia, F., R. Chatterjee, and J. H. May (2019). “Using conditional re-

stricted Boltzmann machines to model complex consumer shopping
patterns”. Marketing Science. 38(4): 711–727.

Zhang, M. and L. Luo (2023). “Can consumer-posted photos serve

as a leading indicator of restaurant survival? Evidence from Yelp”.
Management Science. 69(1): 25–50.
 Full text available at: http://dx.doi.org/10.1561/1700000080

References

Abadie, A., A. Diamond, and J. Hainmueller (2010). “Synthetic con-

trol methods for comparative case studies: Estimating the effect
of California’s tobacco control program”. Journal of the American
Statistical Association. 105(490): 493–505.
Abadie, A. and J. Gardeazabal (2003). “The economic costs of conflict:
A case study of the Basque Country”. American Economic Review.
93(1): 113–132.
Abadie, A. and G. W. Imbens (2006). “Large sample properties of
matching estimators for average treatment effects”. Econometrica.
74(1): 235–267.
Abadie, A. and G. W. Imbens (2011). “Bias-corrected matching estima-
tors for average treatment effects”. Journal of Business & Economic
Statistics. 29(1): 1–11.
Adalja, A., J. Liaukonyte, E. Wang, and X. Zhu (2023). “GMO and non-
GMO labeling effects: Evidence from a quasi-natural experiment”.
Marketing Science. 42(2): 233–250.
Anand, P. and V. Kadiyali (2024). “Frontiers: Smoke and mirrors:
Impact of e-cigarette taxes on underage social media posting”. Mar-
keting Science. doi: 10.1287/mksc.2022.0241.
Angrist, J. D. and J.-S. Pischke (2009). Mostly Harmless Econometrics:
An Empiricist’s Companion. Princeton University Press.

122
 Full text available at: http://dx.doi.org/10.1561/1700000080

References 123

Araya, S., A. Elberg, C. Noton, and D. Schwartz (2022). “Identifying

food labeling effects on consumer behavior”. Marketing Science.
41(5): 982–1003.
Archak, N., A. Ghose, and P. G. Ipeirotis (2011). “Deriving the pricing
power of product features by mining consumer reviews”. Manage-
ment Science. 57(8): 1485–1509.
Athey, S. and G. Imbens (2016). “Recursive partitioning for hetero-
geneous causal effects”. Proceedings of the National Academy of
Sciences. 113(27): 7353–7360.
Athey, S. and G. W. Imbens (2015). “Machine learning methods for
estimating heterogeneous causal effects”. Stat. 1050(5): 1–26.
Athey, S. and G. W. Imbens (2019). “Machine learning methods that
economists should know about”. Annual Review of Economics. 11:
685–725.
Athey, S. and G. W. Imbens (2022). “Design-based analysis in difference-
in-differences settings with staggered adoption”. Journal of Econo-
metrics. 226(1): 62–79.
Athey, S., G. W. Imbens, and S. Wager (2018). “Approximate resid-
ual balancing: Debiased inference of average treatment effects in
high dimensions”. Journal of the Royal Statistical Society Series B:
Statistical Methodology. 80(4): 597–623.
Athey, S. and S. Wager (2019). “Estimating treatment effects with
causal forests: An application”. Observational Studies. 5(2): 37–51.
Bach, P., V. Chernozhukov, M. S. Kurz, and M. Spindler (2021).
“DoubleML–an object-oriented implementation of double machine
learning in R”. arXiv preprint arXiv: 2103.09603.
Bach, P., V. Chernozhukov, M. S. Kurz, and M. Spindler (2022). “Dou-
bleML: An object-oriented implementation of double machine learn-
ing in Python”. The Journal of Machine Learning Research. 23(1):
2469–2474.
Bai, J. (2003). “Inferential theory for factor models of large dimensions”.
Econometrica. 71(1): 135–171.
Bai, J. (2009). “Panel data models with interactive fixed effects”. Econo-
metrica. 77(4): 1229–1279.
Bai, J. and S. Ng (2002). “Determining the number of factors in ap-
proximate factor models”. Econometrica. 70(1): 191–221.
 Full text available at: http://dx.doi.org/10.1561/1700000080

124 References

Baker, A. C., D. F. Larcker, and C. C. Wang (2022). “How much should

we trust staggered difference-in-differences estimates?” Journal of
Financial Economics. 144(2): 370–395.
Barron, K., E. Kung, and D. Proserpio (2021). “The effect of home-
sharing on house prices and rents: Evidence from Airbnb”. Marketing
Science. 40(1): 23–47.
Bettman, J. R. (1979). An Information Processing Theory of Consumer
Choice. Addison-Wesley.
Blei, D. M., A. Y. Ng, and M. I. Jordan (2003). “Latent Dirichlet
allocation”. Journal of Machine Learning Research. 3(Jan): 993–
1022.
Borusyak, K. and P. Hull (2020). “Non-random exposure to exogenous
shocks: Theory and applications”. Tech. Rep. National Bureau of
Economic Research.
Bound, J., D. A. Jaeger, and R. M. Baker (1995). “Problems with
instrumental variables estimation when the correlation between
the instruments and the endogenous explanatory variable is weak”.
Journal of the American Statistical Association. 90(430): 443–450.
Calvo, E., R. Cui, and J. C. Serpa (2019). “Oversight and efficiency in
public projects: A regression discontinuity analysis”. Management
Science. 65(12): 5651–5675.
Campbell, R. H. (1965). “A managerial approach to advertising mea-
surement”. Journal of Marketing. 29(4): 1–6.
Card, D. and A. B. Krueger (1993). “Minimum wages and employment:
A case study of the fast food industry in New Jersey and Pennsyl-
vania”. NBER Working Paper 4509. National Bureau of Economic
Research.
Chen, W., Z. Wei, and K. Xie (2022). “The battle for homes: How
does home sharing disrupt local residential markets?” Management
Science. 68(12): 8589–8612.
Chernozhukov, V., D. Chetverikov, M. Demirer, E. Duflo, C. Hansen, W.
Newey, and J. Robins (2018a). “Double/debiased machine learning
for treatment and structural parameters”. The Econometrics Journal.
21(1): C1–C68. doi: 10.1111/ectj.12097.
 Full text available at: http://dx.doi.org/10.1561/1700000080

References 125

Chernozhukov, V., M. Demirer, E. Duflo, and I. Fernandez-Val (2018b).

“Generic machine learning inference on heterogeneous treatment
effects in randomized experiments, with an application to immuniza-
tion in India”. Tech. Rep. National Bureau of Economic Research.
Chesnes, M., W. Dai, and G. Zhe Jin (2017). “Banning foreign pharma-
cies from sponsored search: The online consumer response”. Market-
ing Science. 36(6): 879–907.
Cochean, W. G. and G. M. Cox (1957). “Experimental designs”. Soil
Science. 84(5): 425.
Cochran, W. G. (1968). “The effectiveness of adjustment by subclassifi-
cation in removing bias in observational studies”. Biometrics: 295–
313.
Cook, T. D. and D. T. Campbell (1979). Quasi-Experimentation: Design
& Analysis Issues for Field Settings. Boston: Houghton Mifflin.
Cui, P. and S. Athey (2022). “Stable learning establishes some common
ground between causal inference and machine learning”. Nature
Machine Intelligence. 4(2): 110–115.
Culotta, A. and J. Cutler (2016). “Mining brand perceptions from
twitter social networks”. Marketing Science. 35(3): 343–362.
Datta, H., G. Knox, and B. J. Bronnenberg (2018). “Changing their
tune: How consumers’ adoption of online streaming affects music
consumption and discovery”. Marketing Science. 37(1): 5–21.
Daviet, R., G. Nave, and J. Wind (2022). “Genetic data: Potential uses
and misuses in marketing”. Journal of Marketing. 86(1): 7–26.
De Haan, E., P. Kannan, P. C. Verhoef, and T. Wiesel (2018). “Device
switching in online purchasing: Examining the strategic contingen-
cies”. Journal of Marketing. 82(5): 1–19.
Doudchenko, N. and G. W. Imbens (2016). “Balancing, regression,
difference-in-differences and synthetic control methods: A synthesis”.
Tech. Rep. National Bureau of Economic Research.
Dzyabura, D., S. El Kihal, J. R. Hauser, and M. Ibragimov (2023).
“Leveraging the power of images in managing product return rates”.
Marketing Science. 42(6): 1125–1142.
Eggert, A., L. Steinhoff, and C. Witte (2019). “Gift purchases as cata-
lysts for strengthening customer-brand relationships”. Journal of
Marketing. 83(5): 115–132.
 Full text available at: http://dx.doi.org/10.1561/1700000080

126 References

Ellickson, P. B., W. Kar, and J. C. Reeder III (2023). “Estimating

marketing component effects: Double machine learning from targeted
digital promotions”. Marketing Science. 42(4): 704–728.
Farrell, M. H., T. Liang, and S. Misra (2021). “Deep neural networks
for estimation and inference”. Econometrica. 89(1): 181–213.
Feder, A., K. A. Keith, E. Manzoor, R. Pryzant, D. Sridhar, Z. Wood-
Doughty, J. Eisenstein, J. Grimmer, R. Reichart, M. E. Roberts,
B. M. Stewart, V. Veitch, and D. Yang (2022). “Causal inference in
natural language processing: Estimation, prediction, interpretation
and beyond”. Transactions of the Association for Computational
Linguistics. 10: 1138–1158.
Fisher, R. A. (1925). Statistical Methods for Research Workers. Oliver
and Boyd.
Fisher, R. A. (1935). “Design of experiments”. British Medical Journal.
1(3923): 554.
Flammer, C. (2015). “Does corporate social responsibility lead to su-
perior financial performance? A regression discontinuity approach”.
Management Science. 61(11): 2549–2568.
Fong, J. and M. Hunter (2022). “Can facing the truth improve outcomes?
Effects of information in consumer finance”. Marketing Science.
41(1): 33–50.
Goldfarb, A., C. Tucker, and Y. Wang (2022). “Conducting research
in marketing with quasi-experiments”. Journal of Marketing. 86(3):
1–20.
Guo, T., S. Sriram, and P. Manchanda (2020). “‘Let the sunshine in’:
The impact of industry payment disclosure on physician prescription
behavior”. Marketing Science. 39(3): 516–539.
Gupta, S. and D. Lehmann (2005). Managing Customers as Investments
the Strategic Value of Customers in the Long Run. Wharton School
Publishing.
Hahn, J., P. Todd, and W. Van der Klaauw (2001). “Identification
and estimation of treatment effects with a regression-discontinuity
design”. Econometrica. 69(1): 201–209.
Hartford, J., G. Lewis, K. Leyton-Brown, and M. Taddy (2017). “Deep
IV: A flexible approach for counterfactual prediction”. In: Interna-
tional Conference on Machine Learning. PMLR. 1414–1423.
 Full text available at: http://dx.doi.org/10.1561/1700000080

References 127

Hartmann, J. and O. Netzer (2023). “Natural language processing in

marketing”. In: Artificial Intelligence in Marketing. Vol. 20. Emerald
Publishing Limited. 191–215.
Hartmann, W., H. S. Nair, and S. Narayanan (2011). “Identifying
causal marketing mix effects using a regression discontinuity design”.
Marketing Science. 30(6): 1079–1097.
Hausman, C. and D. S. Rapson (2018). “Regression discontinuity in
time: Considerations for empirical applications”. Annual Review of
Resource Economics. 10: 533–552.
Henderson, D. J. and S. Sperlich (2023). “A complete framework for
model-free difference-in-differences estimation”. Foundations and
Trends® in Econometrics. 12(3): 232–323.
Hernán, M. and J. Robins (2020). Causal Inference: What If. Boca
Raton, FL: Chapman & Hall/CRC.
Howard, J. A. and J. N. Sheth (1969). The Theory of Buyer Behavior.
Vol. 63. New York. 145.
Hsiao, C., H. Steve Ching, and S. Ki Wan (2012). “A panel data
approach for program evaluation: Measuring the benefits of political
and economic integration of Hong Kong with mainland China”.
Journal of Applied Econometrics. 27(5): 705–740.
Huang, Q., V. R. Nijs, K. Hansen, and E. T. Anderson (2012). “Wal-
Mart’s impact on supplier profits”. Journal of Marketing Research.
49(2): 131–143.
Hui, S. K., J. J. Inman, Y. Huang, and J. Suher (2013). “The effect
of in-store travel distance on unplanned spending: Applications to
mobile promotion strategies”. Journal of Marketing. 77(2): 1–16.
Imbens, G. W. and T. Lemieux (2008). “Regression discontinuity designs:
A guide to practice”. Journal of Econometrics. 142(2): 615–635.
Imbens, G. W. and D. B. Rubin (2015). Causal Inference in Statistics,
Social, and Biomedical Sciences. Cambridge University Press.
Iyengar, R., Y.-H. Park, and Q. Yu (2022). “The impact of subscription
programs on customer purchases”. Journal of Marketing Research.
59(6): 1101–1119.
 Full text available at: http://dx.doi.org/10.1561/1700000080

128 References

Jo, W., S. Sunder, J. Choi, and M. Trivedi (2020). “Protecting consumers

from themselves: Assessing consequences of usage restriction laws
on online game usage and spending”. Marketing Science. 39(1): 117–
133.
Johansson, F., U. Shalit, and D. Sontag (2016). “Learning representa-
tions for counterfactual inference”. In: International Conference on
Machine Learning. PMLR. 3020–3029.
Joo, M., D. K. Gauri, and K. C. Wilbur (2020). “Temporal distance and
price responsiveness: Empirical investigation of the cruise industry”.
Management Science. 66(11): 5362–5388.
Kalisch, M. and P. Bühlman (2007). “Estimating high-dimensional
directed acyclic graphs with the PC-algorithm”. Journal of Machine
Learning Research. 8(3): 613–636.
Kim, S., C. Lee, and S. Gupta (2020). “Bayesian synthetic control
methods”. Journal of Marketing Research. 57(5): 831–852.
Kotler, P. and K. L. Keller (2012). Marketing Management. Pearson.
Krijestorac, H., R. Garg, and V. Mahajan (2020). “Cross-platform
spillover effects in consumption of viral content: A quasi-experimental
analysis using synthetic controls”. Information Systems Research.
31(2): 449–472.
Kumar, V. (2013). Profitable Customer Engagement: Concept, Metrics
and Strategies. India: SAGE Publications.
Lavidge, R. J. and G. A. Steiner (1961). “A model for predictive mea-
surements of advertising effectiveness”. Journal of Marketing. 25(6):
59–62.
Le, T. D., T. Hoang, J. Li, L. Liu, H. Liu, and S. Hu (2016). “A fast
PC algorithm for high dimensional causal discovery with multi-
core PCs”. IEEE/ACM Transactions on Computational Biology and
Bioinformatics. 16(5): 1483–1495.
Lee, B. K., J. Lessler, and E. A. Stuart (2010). “Improving propensity
score weighting using machine learning”. Statistics in Medicine.
29(3): 337–346.
Lee, D. S. and T. Lemieux (2010). “Regression discontinuity designs in
economics”. Journal of Economic Literature. 48(2): 281–355.
 Full text available at: http://dx.doi.org/10.1561/1700000080

References 129

Lee, D., K. Hosanagar, and H. S. Nair (2018). “Advertising content and

consumer engagement on social media: Evidence from Facebook”.
Management Science. 64(11): 5105–5131.
Lee, T. Y. and E. T. Bradlow (2011). “Automated marketing research
using online customer reviews”. Journal of Marketing Research.
48(5): 881–894.
Li, K. T. (2023). “Frontiers: A simple forward difference-in-differences
method”. Marketing Science. doi: 10.1287/mksc.2022.0212.
Li, K. T. and D. R. Bell (2017). “Estimation of average treatment effects
with panel data: Asymptotic theory and implementation”. Journal
of Econometrics. 197(1): 65–75.
Li, K. T. and V. Shankar (2023). “A two-step synthetic control approach
for estimating causal effects of marketing events”. Management
Science. doi: 10.1287/mnsc.2023.4878.
Li, K. T. and G. P. Sonnier (2023). “Statistical inference for the factor
model approach to estimate causal effects in quasi-experimental
settings”. Journal of Marketing Research. 60(3): 449–472.
Li, K. T. and C. Van den Bulte (2023). “Augmented difference-in-
differences”. Marketing Science. 42(4): 746–767.
Li, L., S. Gopinath, and S. J. Carson (2022). “History matters: The
impact of online customer reviews across product generations”.
Management Science. 68(5): 3878–3903.
Li, Q. and N. Lin (2010). “The Bayesian elastic net”. Bayesian Analysis.
5(1): 151–170.
Li, X., C. Liao, and Y. Xie (2021). “Digital piracy, creative productiv-
ity, and customer care effort: Evidence from the digital publishing
industry”. Marketing Science. 40(4): 685–707.
Li, X., M. Shi, and X. S. Wang (2019). “Video mining: Measuring visual
information using automatic methods”. International Journal of
Research in Marketing. 36(2): 216–231.
Liu, X., D. Lee, and K. Srinivasan (2019). “Large-scale cross-category
analysis of consumer review content on sales conversion leveraging
deep learning”. Journal of Marketing Research. 56(6): 918–943.
 Full text available at: http://dx.doi.org/10.1561/1700000080

130 References

Ludwig, S., K. De Ruyter, M. Friedman, E. C. Brüggen, M. Wetzels,

and G. Pfann (2013). “More than words: The influence of affective
content and linguistic style matches in online reviews on conversion
rates”. Journal of Marketing. 77(1): 87–103.
Manchanda, P., G. Packard, and A. Pattabhiramaiah (2015). “Social
dollars: The economic impact of customer participation in a firm-
sponsored online customer community”. Marketing Science. 34(3):
367–387.
McCrary, J. (2008). “Manipulation of the running variable in the regres-
sion discontinuity design: A density test”. Journal of Econometrics.
142(2): 698–714.
Montaguti, E., S. A. Neslin, and S. Valentini (2016). “Can market-
ing campaigns induce multichannel buying and more profitable
customers? A field experiment”. Marketing Science. 35(2): 201–217.
Moshary, S., B. T. Shapiro, and J. Song (2021). “How and when to use
the political cycle to identify advertising effects”. Marketing Science.
40(2): 283–304.
Narang, U. and V. Shankar (2019). “Mobile app introduction and online
and offline purchases and product returns”. Marketing Science. 38(5):
756–772.
Narayanan, S. and K. Kalyanam (2015). “Position effects in search ad-
vertising and their moderators: A regression discontinuity approach”.
Marketing Science. 34(3): 388–407.
Netzer, O., R. Feldman, J. Goldenberg, and M. Fresko (2012). “Mine
your own business: Market-structure surveillance through text min-
ing”. Marketing Science. 31(3): 521–543.
Neyman, J. (1923, 1990). “On the application of probability theory to
agricultural experiments. Essay on principles”. Ann. Agricultural
Sciences: 1–51.
Oblander, S. and D. M. McCarthy (2023). “Frontiers: Estimating the
long-term impact of major events on consumption patterns: Evidence
from COVID-19”. Marketing Science. 42(5): 839–852.
Oster, E. (2019). “Unobservable selection and coefficient stability: The-
ory and evidence”. Journal of Business & Economic Statistics. 37(2):
187–204.
 Full text available at: http://dx.doi.org/10.1561/1700000080

References 131

Ozturk, O. C., P. K. Chintagunta, and S. Venkataraman (2019). “Con-

sumer response to Chapter 11 bankruptcy: Negative demand spillover
to competitors”. Marketing Science. 38(2): 296–316.
Pattabhiramaiah, A., E. Overby, and L. Xu (2022). “Spillovers from
online engagement: How a newspaper subscriber’s activation of
digital paywall access affects her retention and subscription revenue”.
Management Science. 68(5): 3528–3548.
Pearl, J. (1995). “Causal diagrams for empirical research”. Biometrika.
82(4): 669–688.
Pearl, J. (2009a). Causality: Models, Reasoning and Inference. 2nd edn.
Cambridge University Press.
Pearl, J. (2009b). “Myth, confusion, and science in causal analysis”.
Tech. Rep. R-348. University of California, LA.
Pearl, J. and D. Mackenzie (2018). The Book of Why: The New Science
of Cause and Effect. Basic Books.
Puranam, D., V. Kadiyali, and V. Narayan (2021). “The impact of
increase in minimum wages on consumer perceptions of service: A
transformer model of online restaurant reviews”. Marketing Science.
40(5): 985–1004.
Puranam, D., V. Narayan, and V. Kadiyali (2017). “The effect of calorie
posting regulation on consumer opinion: A flexible latent Dirichlet
allocation model with informative priors”. Marketing Science. 36(5):
726–746.
Rafieian, O. and H. Yoganarasimhan (2021). “Targeting and privacy in
mobile advertising”. Marketing Science. 40(2): 193–218.
Rambocas, M. and B. G. Pacheco (2018). “Online sentiment analysis in
marketing research: A review”. Journal of Research in Interactive
Marketing. 12(2): 146–163.
Rosenbaum, P. R. and D. B. Rubin (1983). “The central role of the
propensity score in observational studies for causal effects”. Biomet-
rika. 70(1): 41–55.
Rossi, P. E. (2014). “Even the rich can make themselves poor: A critical
examination of IV methods in marketing applications”. Marketing
Science. 33(5): 655–672.
 Full text available at: http://dx.doi.org/10.1561/1700000080

132 References

Rubin, D. B. (1974). “Estimating causal effects of treatments in random-

ized and nonrandomized studies”. Journal of Educational Psychology.
66(5): 688–701.
Rubin, D. B. (1980). “Randomization analysis of experimental data:
The Fisher randomization test comment”. Journal of the American
Statistical Association. 75(371): 591–593.
Rutz, O. J., R. E. Bucklin, and G. P. Sonnier (2012). “A latent instru-
mental variables approach to modeling keyword conversion in paid
search advertising”. Journal of Marketing Research. 49(3): 306–319.
Sahoo, N., C. Dellarocas, and S. Srinivasan (2018). “The impact of
online product reviews on product returns”. Information Systems
Research. 29(3): 723–738.
Seiler, S., S. Yao, and W. Wang (2017). “Does online word of mouth
increase demand? (And how?) Evidence from a natural experiment”.
Marketing Science. 36(6): 838–861.
Shalit, U., F. D. Johansson, and D. Sontag (2017). “Estimating indi-
vidual treatment effect: generalization bounds and algorithms”. In:
International Conference on Machine Learning. PMLR. 3076–3085.
Shen, Z., P. Cui, T. Zhang, and K. Kunag (2020). “Stable learning
via sample reweighting”. Proceedings of the AAAI Conference on
Artificial Intelligence. 34(4): 5692–5699.
Simonov, A., S. Sacher, J.-P. Dubé, and S. Biswas (2022). “Frontiers: the
persuasive effect of fox news: Noncompliance with social distancing
during the COVID-19 pandemic”. Marketing Science. 41(2): 230–
242.
Sonnier, G. P., L. McAlister, and O. J. Rutz (2011). “A dynamic model
of the effect of online communications on firm sales”. Marketing
Science. 30(4): 702–716.
Spirtes, P. and C. Glymour (1991). “An algorithm for fast recovery of
sparse causal graphs”. Social Science Computer Review. 9(1): 62–72.
Spirtes, P., C. N. Glymour, and R. Scheines (2000). Causation, Predic-
tion, and Search. MIT Press.
Sudhir, K. and D. Talukdar (2015). “The “Peter Pan syndrome” in
emerging markets: The productivity-transparency trade-off in IT
adoption”. Marketing Science. 34(4): 500–521.
 Full text available at: http://dx.doi.org/10.1561/1700000080

References 133

Tan, A.-H. (1999). “Text mining: The state of the art and the challenges”.
Proceedings of the PAKDD 1999 Workshop on Knowledge Discovery
from Advanced Databases. 8: 65–70.
Thistlethwaite, D. L. and D. T. Campbell (1960). “Regression-discon-
tinuity analysis: An alternative to the ex post facto experiment”.
Journal of Educational Psychology. 51(6): 309.
Thomas, M. (2020). “Spillovers from mass advertising: An identification
strategy”. Marketing Science. 39(4): 807–826.
Tian, Z., R. Dew, and R. Iyengar (2023). “Mega or Micro? Influencer
selection using follower elasticity”. Journal of Marketing Research.
doi: 10.1177/00222437231210267.
Tirunillai, S. and G. J. Tellis (2017). “Does offline TV advertising affect
online chatter? Quasi-experimental analysis using synthetic control”.
Marketing Science. 36(6): 862–878.
Unal, M. and Y.-H. Park (2023). “Fewer clicks, more purchases”. Man-
agement Science. 69(12): 7317–7334.
Wager, S. and S. Athey (2018). “Estimation and inference of hetero-
geneous treatment effects using random forests”. Journal of the
American Statistical Association. 113(523): 1228–1242.
Wang, Y., M. S. Qin, X. Luo, and Y. Kou (2022). “Frontiers: How
support for Black Lives Matter impacts consumer responses on
social media”. Marketing Science. 41(6): 1029–1044.
Wickens, C. D. and S. R. Dixon (2007). “The benefits of imperfect
diagnostic automation: A synthesis of the literature”. Theoretical
Issues in Ergonomics Science. 8(3): 201–212.
Xu, Y. (2017). “Generalized synthetic control method: Causal inference
with interactive fixed effects models”. Political Analysis. 25(1): 57–
76.
Yan, S., K. M. Miller, and B. Skiera (2022). “How does the adoption
of ad blockers affect news consumption?” Journal of Marketing
Research. 59(5): 1002–1018.
Yazdani, E., S. Gopinath, and S. Carson (2018). “Preaching to the choir:
The chasm between top-ranked reviewers, mainstream customers,
and product sales”. Marketing Science. 37(5): 838–851.
 Full text available at: http://dx.doi.org/10.1561/1700000080

134 References

Zhong, Z. (2022). “Chasing diamonds and crowns: Consumer limited

attention and seller response”. Management Science. 68(6): 4380–
4397.
Zubizarreta, J. R. (2015). “Stable weights that balance covariates for
estimation with incomplete outcome data”. Journal of the American
Statistical Association. 110(511): 910–922.