Vamsi Manthena

Vamsi Manthena

Data Scientist II | Ph.D. Statistics

Expedia Group

“It’s easy to lie with statistics, it’s hard to tell the truth without statistics.”


Statistics allows us to answer the questions of today to pursue a brighter tomorrow. Data helps us break away from old mindsets and opens up new possibilities. Statistics has the power to improve every aspect of human life. This sums up my approach statistics and data science.

I love chatting about all things data, statistics, and life. Connect with me at vamsi.manthena@gmail.com.

Interests
  • Experimentation and A/B Testing
  • Causal Inference
  • Variance Reduction
  • Deep Learning (GANs)
  • Machine Learning
  • Data Visualization
Education

  • PhD Statistics, 2022

    University of Nebraska - Lincoln

  • MS Statistics, 2019

    University of Nebraska - Lincoln

  • MSc Mathematical Sciences, 2016

    University of Hyderabad, India

Skills

Python
R
Spark/Scala
SQL
Machine Learning
Data Visualization

Publications

Full list @ Google Scholar

Integrating and optimizing genomic, weather, and secondary trait data for multiclass classification
This paper examines the spatio-temporal dynamics of a marine ecosystem using reaction–diffusion equations.
PDF Cite
Integrating and optimizing genomic, weather, and secondary trait data for multiclass classification
Evaluating dimensionality reduction for genomic prediction
This paper explored the effectiveness of dimensionality reduction as a pre-processing step for genomic prediction
PDF Cite
Evaluating dimensionality reduction for genomic prediction
Modelling and analysis of spatio-temporal dynamics of a marine ecosystem
This paper examines the spatio-temporal dynamics of a marine ecosystem using reaction–diffusion equations.
PDF Cite
Modelling and analysis of spatio-temporal dynamics of a marine ecosystem

Projects

*
All Research Data Visualization Personal Projects
Combining Multi-type Data to Improve Genomic Prediction

Combining Multi-type Data to Improve Genomic Prediction

Novel methods to combine multi-type data to improve high-dimensional classification.

Slides Video
The Search for Optimal Weather Window

The Search for Optimal Weather Window

Determining the optimal weather window to improve multi-type genomic prediction

Slides
Evaluating Dimensionality Reduction for Genomic Prediction

Evaluating Dimensionality Reduction for Genomic Prediction

Learn how dimensionality reduction methods can be used to improve genomic prediction models.

Slides Video
Liver Disease Prediction using Python

Liver Disease Prediction using Python

Learn boosting methods and python using liver data.

Code Dataset
2020 US Presidential Election

2020 US Presidential Election

Who would have won the 2020 US election if only ______ group voted? Find out using the interactive dashboard here .

Dataset Dashboard Link
Hello, Tableau!

Hello, Tableau!

How safe are London buses? Find out using the interactive dashboard here .

Dataset Dashboard Link
Coal Production in India (2019-2020)

Coal Production in India (2019-2020)

Learn more about coal production using the interactive dashboard here .

Dataset Dashboard Link

Experience

 
 
 
 
 
Data Scientist II
Expedia Group
Jun 2022 – Present Seattle, WA
Conducted end-to-end experimentation for travel ads through test designing, test sizing, defining success criteria, and hypothesis testing. Responsible for implementing CUPED and other ML variance reduction methods to improve A/B testing velocity. Implemented both A/B tests and causal inference methods to measure the impact of feature changes on travel ads metrics. Supervised and mentored 3 colleagues through various analytics projects.
 
 
 
 
 
Statistical Consultant
Statistical Cross-disciplinary Collboration and Consulting Lab (SC3L)
May 2020 – May 2022 Lincoln, Nebraska
Collaborated with cross-disciplinary researchers on over 25 projects across the university and assisted them with experimental design, power analyses, summarizing data, statistical analyses, result interpretation, and communication using SAS and R. Prepared reports and explained complex statistical methods to subject matter experts. Applied methods such as generalized linear mixed models, longitudinal analysis, regression and classification.
 
 
 
 
 
Statistics Intern
Centre for Brain, Biology and Behavior
May 2020 – Aug 2021 Lincoln, Nebraska
Developing and testing methods to predict psychiatric outcomes from high-dimensional MRI data using Python and SAS. Implemented methods such as random forest, LCAs, SVMs, boosting, clustering, and network centrality and stability metrics using Python and R.
 
 
 
 
 
Graduate Research Assistant
University of Nebraska - Lincoln
Aug 2019 – May 2020 Lincoln, Nebraska
Evaluated dimensionality reduction methods such as clustering, randomized sketching, deterministic sampling, and shrinkage methods in the genomic prediction problem using R and HPC (high-performance computing). Achieved the full data prediction accuracy using only 10% of the data.
 
 
 
 
 
Sports Analytics Intern
Nebraska Sports Analytics
Sep 2017 – Dec 2019 Lincoln, Nebraska
Investigated the effect of scheduling and workload on the win-probability for the Big Ten men’s basketball teams based on models such as GLMs, tree based methods, and neural networks. Collected, cleaned, and compiled data from public datasets using R.
 
 
 
 
 
Instructor of Record
University of Nebraska - Lincoln
Aug 2017 – Aug 2019 Lincoln, Nebraska

STAT 380: Statistics and Applications

  • Instructor of record (Summer 2018, Spring 2019, Summer 2019), Co-instructor (Fall 2017, Spring 2018).
  • Calculus based introductory statistics course.
  • Primary focus was on probability theory and introduction to statistical inference.

STAT 218: Introduction to Statistics

  • Instructor of record (Fall 2018).
  • Non-calculus based introductory statistics course.
  • Primary focus was on statistical literacy and statistical communication.

Contact

Reach out to me to chat about data, statistics and my work!