Mukul Pareek

  • Adjunct Professor

Connect

Mukul Pareek headshot

Research Interests: Cybersecurity, Data Science

Mukul is an industry professional with experience in data, analytics, audit, accounting, risk management and cybersecurity.  


Education:

1. MBA, Columbia Business School and London Business School (EMBA Global Program), 2003

2. B.Com (Hons.), Shriram College of Commerce, University of Delhi, 1990

 

Industry Credentials:

1. Chartered Accountant, India, 1993

2. Cost Management Accountant, India, 1990

3. Certified Information Systems Auditor (ISACA), 1997

4. Professional Risk Manager (PRMIA), 2009

 


I teach Business Analytics.  So what does that mean?  We cover the below syllabus.  We also cover six case studies, and do a final project on a real data set.

Class 1: Introduction
Overview of analytics in business, the nature of problems solved for and a brief recap of descriptive statistics with Python.

Class 2: Exploratory Data Analysis
Exploratory data analysis, data summaries, basic graphing, pivots, data profiling etc.

Class 3: Data visualization
Visualization with Python/Matplotlib covering major graph types - scatterplot, histograms and barplots, boxplots, pairplots etc.

Class 4: Data cleaning and transformations
Managing data - dealing with missing values, transforming data for analytics.

Class 5: Introduction to modeling
A gentle introduction to modeling, covering key concepts relating to approaches, the machine learning workflow and model evaluation.

Class 6: Regression
A conceptual review of linear regression (including ridge and lasso regularization), polynomial regression, logistic regression, and the nature of problems these techniques can address.

Class 7: Feature engineering
The importance of feature engineering and key transformations that help transform data to useful features.

Class 8: Modeling and machine learning - Part I
The machine learning process.  We will learn the difference between supervised and unsupervised learning, tree methods, linear discriminants and SVMs etc, training and test sets, and model evaluation.

Class 9: Modeling and machine learning - Part II
We will continue the discussion from the prior class and complete our review of key algorithms.

Class 10: Deep learning
A review of neural networks, and the nature of problems they solve for.  We will cover at a high level fully connected neural nets, RNNs, CNNs, backpropagation and other key concepts.

Class 11: Text data
Approaches to dealing with text data.  Word counts, bag-of-word approaches, term matrices, and a high level discussion of vectorization techniques such as Glove, Word2Vec and BERT

Class 12: Time series analysis
Analyzing and forecasting time series data.  We will cover practical applications of ETS decomposition, Holt-Winters, SARIMA with a view to addressing business problems.

Class 13: Excel as a business tool
Excel as a business analytics tool. Best practices for use and sharing.

Class 14: Organizational considerations
Getting analytics into production, maintaining sponsorship, managing your manager, explainability & story telling, managing change, building teams, and operating effectively.