Selected Publications

We propose hardware lifecycle as a key moderator of the impact of superstar and non-superstar software on hardware adoption. A hardware’s earlier adopters, are less price sensitive and have a higher preference for exciting and challenging software. In contrast, later adopters are more price sensitive and prefer simplicity in software. Superstars tend to be more expensive and more complex compared to non-superstars. Therefore, earlier (later) adopters prefer superstars (non- superstars), which leads to higher impact of superstars (non-superstars) on hardware adoption in the early (later) stages of the hardware lifecycle. Using monthly data over a twelve-year timeframe (1995 – 2007) from the home video game industry, we find that both superstar and non-superstar software impact hardware demand, but they matter at different points in the hardware lifecycle. Superstars are most influential when hardware is new and this influence declines as hardware ages. In contrast, non-superstar software has a positive impact on hardware demand later in the hardware lifecycle, and this impact increases with hardware age. Findings reveal that eventually the amount of available non-superstar software impacts hardware adoption more than the amount of available superstar software. The authors provide several managerial implications based on these findings.
In JAMS, 2019

The purpose of this paper is to investigate how social media affects US retailers’ customer-based brand equity (CBBE) which is an important indicator of brand success. The authors find strong impacts of owned and earned social media on CBBE across the board. However, they find that owned social media harms CBBE of retailers dealing in hedonic and high involvement products. Whereas owned social media helps general retailers in building CBBE, it reduces CBBE of specialty retailers.
In Admin Sci., 2018

Companies that engage in corporate activism risk taking stances that reflect the values of their management but alienate key segments of a politically divided customer base. Our data suggests that company executives should think carefully about customers’ political affiliations and likelihood to engage in positive and negative word-of-mouth.
In HBR, 2018

Analyzing daily data for 45 brands in 21 sectors using vector autoregression models, we find that social media brand fan following improves brand awareness, purchase intent, and customer satisfaction. Earned social media engagement volume affects brand awareness and purchase intent but not customer satisfaction, while positive and negative valence have the largest effects on customer satisfaction. Owned social media increases brand awareness and customer satisfaction but not purchase intent, highlighting a nonlinear effect of brand-controlled social media.
In JM, 2018

Recent & Upcoming Talks

Example Talk
Jun 1, 2030 1:00 PM

Recent Posts

More Posts

In this post, we will visualize spread of worldwide COVID-19 cases through time. I obtained the data from Rami Krispin’s website: using coronovirus package. I also decided to do some experimentation using John Coene’s fantastic echarts4r package, which allows us to access echarts API. Load the libraries and get the data in the R session. library(dplyr) library(echarts4r) library(coronavirus) # Get the data data("coronavirus") Data Preparation Check out the first 6 observations.


This post will help you set up RStudio Cloud for the UTSA-GenAI workshop to be held on February 14, 2020. In the workshop I will show you a lesson that involved topic modeling on product reviews using Latent Dirichlet Allocation. The lesson requires a code file, a data file, and a word list. I have set up a R project on RStudio Cloud, which you can start working on right away by following the instructions below.


I came across the visualization of US opioid epidemic made by Kieran Healy in his book “Data Visualization: A Practical Introduction” (Link). He has used data through 2014 but in recent years the epidemic has become worse. So I extended the data to 2017 by downloading it from Kaiser Family Foundation’s website. After cleaning up the data, I ended up with an unbalanced panel of 50 states over 1999 to 2017.


I came across a few nice applications-related posts. Machine Learning Music Composed From Re-Synthesized Fragments From 100s Of Terabytes Of LA Phil Recordings: This posts has “a new high def version of the dazzling 3D video/AI-driven performance displayed on the Walt Disney Concert Hall last year.” AI-driven music is nothing new. About 3 years ago I showed a video of computer algorithm creating fantastic music to my students and some of them became upset!


I came across this Science article someone shared on Twitter: Plastic waste inputs from land in to the ocean In this post I am going to make a bar graph using the top 10 countries listed in Table 1. Here is the screenshot of that table. The ranking in the table is based on the last column. As that column shows interval estimates, I decided use the midpoints of those intervals.



I am scheduled to teach the following courses in fall 2018:

  • DA 6233 Section 1B: Data Analytics Visualization and Communication (MS in Data Analytics)
  • DA 6233 Section 2B: Data Analytics Visualization and Communication (MS in Data Analytics)
  • MKT 7043 Section 2: Seminar in Marketing Strategy (Marketing PhD)