NICOLE L. ROBERTS

Data Scientist

Hi, my name is Nicole and I am a Data Scientist based in New York City. Please feel free to reach out.

About Nicole

I am a business solutions-oriented Data Scientist with cross-domain experience spanning Asset Management, FinTech, Marketing, and Retail. Skilled in Python, R, SQL, AWS, data visualization, Adobe Analytics, and Salesforce with a BS degree in Industrial Engineering from Arizona State University's Ira A. Fulton School of Engineering and a FinTech certificate from Columbia Engineering. My experience includes both traditional machine learning and generative AI models.

I approach problem solving from an innovative and lean, business-centric perspective while often assuming a leadership role on projects due to my strong communication and collaboration skills. I am an out-of-the box thinker looking to grow with an innovative organization populated with a team of likewise passionate, motivated team players.

I enjoy playing detective as a data scientist, making discoveries and telling stories with data to improve quantitative business decision making across a variety of sectors. Sharing these insights, packaging them into non-technical, solutions-oriented language digestible for all stakeholders, is one of the most rewarding aspects of my job.

You can count on me to bring the following to every project:

• Clear, effective communication.
• Excellent problem-solving skills from a big picture, business-oriented perspective.
• Ability to clearly tell stories with data and derive smart optimization strategies.
• Experience in leadership roles, having run a business and managed teams.

I look forward to learning about what I can do for your organization.

You can find my resume here.

Modeling the VIX with LSTM

The CBOE VIX is a real-time index that represents the market’s expectations for the relative strength of near-term price changes or future volatility of the S&P 500 Index (SPX), which is considered the leading indicator of the broad U.S. stock market. Being a forward-looking index, it is constructed using the implied volatilities on S&P 500 index options (SPX options).

Volatility, or how fast prices change, is often seen as a way to gauge market sentiment, and in particular the degree of fear among market participants. Having the ability to gauge this sentiment in advance allows insight into the timing of indicators to help traders execute their market entries and exits.  

This project includes various multivariate, multistep LSTM (long short-term memory neural network) models using SPX options data to forecast the CBOE VIX to improve future market volatility forecasts. Monte Carlo simulation and Facebook Prophet forecasts are included for comparison.

Predicting ROI for Option Contracts

This Python project entails a financial services solution that predicts the outcome of an option contract’s return on investment.

I developed a proprietary machine learning pipeline using a set of option contract variables as features along with, notably, predicted VIX prices (derived from own ‘Modeling the VIX with LSTM’ project model).

This model provides traders confidence in their investment decisions, accurately predicting ROI outcome with up to 77% accuracy and 85% precision.

Due to proprietary technology, this code is currently private. Please contact me for any questions; I am happy to provide further detail on an as-needed basis.

ROI % for select SPY call option contracts, Q1 2022.

Using Random Forest to Better Target Asset Management Prospects

Deep Sea Asset Management is a wealth management firm offering advisory services, financial planning, and retirement planning for individuals.

Their primary goal for this analysis was to effectively target and maximize acquisition of higher-value prospects.

I conducted market research for Deep Sea, including developing a random forest model, to identify and help tailor services, products, and marketing strategies to align with the financial behaviors and needs of higher-income segments. This market research aimed to derive insights as to which demographic characteristics indicate higher value individuals.  Knowing which characteristics are the key features that predict clients are above a certain income threshold will help them formulate effective marketing campaigns that are most likely to resonate with this target segment.

*Deep Sea Asset Management is a hypothetical company.

Field Services Management SQL App

I developed a cloud-based field service management SaaS solution that streamlines internal business processes through both a mobile app and desktop dashboard using Python, Flask, HTML, Jinja2, and SQL. The app’s database system, engineered with PostgreSQL, manages users, jobs, technicians, and app-to-technician messages.

The app supports a wide range of field service industries with interchangeable tools such as an integrated CRM, real-time scheduling and dispatching, expense tracking, invoice creation and billing, and technician client reminders.

With its straightforward interface and uncomplicated approach to workflow management, the app drastically reduces the time it takes dispatchers to assign and manage workload and maintain organized, up-to-date accounting – all with minimal clicks.



Let’s Connect

I look forward to collaborating with you. Please feel free to contact me.